[{"file_date_updated":"2023-02-23T10:30:04Z","date_created":"2023-02-02T20:29:57Z","date_updated":"2023-02-23T10:30:54Z","related_material":{"link":[{"relation":"software","description":"source code","url":"https://github.com/ISTAustria-CVML/FLEA"}]},"author":[{"first_name":"Eugenia B","last_name":"Iofinova","id":"f9a17499-f6e0-11ea-865d-fdf9a3f77117","orcid":"0000-0002-7778-3221","full_name":"Iofinova, Eugenia B"},{"full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","first_name":"Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"department":[{"_id":"ChLa"}],"publisher":"ML Research Press","publication_status":"published","acknowledgement":"The authors would like to thank Bernd Prach, Elias Frantar, Alexandra Peste, Mahdi Nikdan, and Peter Súkeník for their helpful feedback. This research was supported by the Scientific Service Units (SSU) of IST Austria through resources provided by Scientific Computing (SciComp). This publication was made possible by an ETH AI Center postdoctoral fellowship granted to Nikola Konstantinov. Eugenia Iofinova was supported in part by the FWF DK VGSCO, grant agreement number W1260-N35. ","year":"2022","publication_identifier":{"issn":["2835-8856"]},"month":"12","language":[{"iso":"eng"}],"acknowledged_ssus":[{"_id":"ScienComp"}],"project":[{"name":"Vienna Graduate School on Computational Optimization","grant_number":" W1260-N35","_id":"9B9290DE-BA93-11EA-9121-9846C619BF3A"}],"quality_controlled":"1","tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"external_id":{"arxiv":["2106.11732"]},"oa":1,"main_file_link":[{"url":"https://openreview.net/forum?id=XsPopigZXV","open_access":"1"}],"abstract":[{"lang":"eng","text":"Fairness-aware learning aims at constructing classifiers that not only make accurate predictions, but also do not discriminate against specific groups. It is a fast-growing area of\r\nmachine learning with far-reaching societal impact. However, existing fair learning methods\r\nare vulnerable to accidental or malicious artifacts in the training data, which can cause\r\nthem to unknowingly produce unfair classifiers. In this work we address the problem of\r\nfair learning from unreliable training data in the robust multisource setting, where the\r\navailable training data comes from multiple sources, a fraction of which might not be representative of the true data distribution. We introduce FLEA, a filtering-based algorithm\r\nthat identifies and suppresses those data sources that would have a negative impact on\r\nfairness or accuracy if they were used for training. As such, FLEA is not a replacement of\r\nprior fairness-aware learning methods but rather an augmentation that makes any of them\r\nrobust against unreliable training data. We show the effectiveness of our approach by a\r\ndiverse range of experiments on multiple datasets. Additionally, we prove formally that\r\n–given enough data– FLEA protects the learner against corruptions as long as the fraction of\r\naffected data sources is less than half. Our source code and documentation are available at\r\nhttps://github.com/ISTAustria-CVML/FLEA."}],"type":"journal_article","oa_version":"Published Version","file":[{"relation":"main_file","file_id":"12673","checksum":"97c8a8470759cab597abb973ca137a3b","success":1,"date_created":"2023-02-23T10:30:04Z","date_updated":"2023-02-23T10:30:04Z","access_level":"open_access","file_name":"2022_TMLR_Iofinova.pdf","content_type":"application/pdf","file_size":1948063,"creator":"dernst"}],"ddc":["000"],"status":"public","title":"FLEA: Provably robust fair multisource learning from unreliable training data","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"12495","article_processing_charge":"No","has_accepted_license":"1","day":"22","date_published":"2022-12-22T00:00:00Z","article_type":"original","citation":{"apa":"Iofinova, E. B., Konstantinov, N. H., & Lampert, C. (2022). FLEA: Provably robust fair multisource learning from unreliable training data. Transactions on Machine Learning Research. ML Research Press.","ieee":"E. B. Iofinova, N. H. Konstantinov, and C. Lampert, “FLEA: Provably robust fair multisource learning from unreliable training data,” Transactions on Machine Learning Research. ML Research Press, 2022.","ista":"Iofinova EB, Konstantinov NH, Lampert C. 2022. FLEA: Provably robust fair multisource learning from unreliable training data. Transactions on Machine Learning Research.","ama":"Iofinova EB, Konstantinov NH, Lampert C. FLEA: Provably robust fair multisource learning from unreliable training data. Transactions on Machine Learning Research. 2022.","chicago":"Iofinova, Eugenia B, Nikola H Konstantinov, and Christoph Lampert. “FLEA: Provably Robust Fair Multisource Learning from Unreliable Training Data.” Transactions on Machine Learning Research. ML Research Press, 2022.","short":"E.B. Iofinova, N.H. Konstantinov, C. Lampert, Transactions on Machine Learning Research (2022).","mla":"Iofinova, Eugenia B., et al. “FLEA: Provably Robust Fair Multisource Learning from Unreliable Training Data.” Transactions on Machine Learning Research, ML Research Press, 2022."},"publication":"Transactions on Machine Learning Research"},{"department":[{"_id":"GradSch"},{"_id":"ChLa"}],"publisher":"Springer Nature","publication_status":"published","year":"2022","volume":13681,"date_updated":"2023-05-03T08:00:46Z","date_created":"2022-08-12T15:09:47Z","author":[{"full_name":"Prach, Bernd","id":"2D561D42-C427-11E9-89B4-9C1AE6697425","last_name":"Prach","first_name":"Bernd"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"quality_controlled":"1","external_id":{"arxiv":["2208.03160"]},"oa":1,"main_file_link":[{"url":" https://doi.org/10.48550/arXiv.2208.03160","open_access":"1"}],"language":[{"iso":"eng"}],"doi":"10.1007/978-3-031-19803-8_21","conference":{"name":"ECCV: European Conference on Computer Vision","end_date":"2022-10-27","start_date":"2022-10-23","location":"Tel Aviv, Israel"},"publication_identifier":{"isbn":["9783031198021"],"eisbn":["9783031198038"]},"month":"10","intvolume":" 13681","title":"Almost-orthogonal layers for efficient general-purpose Lipschitz networks","status":"public","_id":"11839","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","alternative_title":["LNCS"],"type":"conference","abstract":[{"text":"It is a highly desirable property for deep networks to be robust against\r\nsmall input changes. One popular way to achieve this property is by designing\r\nnetworks with a small Lipschitz constant. In this work, we propose a new\r\ntechnique for constructing such Lipschitz networks that has a number of\r\ndesirable properties: it can be applied to any linear network layer\r\n(fully-connected or convolutional), it provides formal guarantees on the\r\nLipschitz constant, it is easy to implement and efficient to run, and it can be\r\ncombined with any training objective and optimization method. In fact, our\r\ntechnique is the first one in the literature that achieves all of these\r\nproperties simultaneously. Our main contribution is a rescaling-based weight\r\nmatrix parametrization that guarantees each network layer to have a Lipschitz\r\nconstant of at most 1 and results in the learned weight matrices to be close to\r\northogonal. Hence we call such layers almost-orthogonal Lipschitz (AOL).\r\nExperiments and ablation studies in the context of image classification with\r\ncertified robust accuracy confirm that AOL layers achieve results that are on\r\npar with most existing methods. Yet, they are simpler to implement and more\r\nbroadly applicable, because they do not require computationally expensive\r\nmatrix orthogonalization or inversion steps as part of the network\r\narchitecture. We provide code at https://github.com/berndprach/AOL.","lang":"eng"}],"page":"350-365","citation":{"short":"B. Prach, C. Lampert, in:, Computer Vision – ECCV 2022, Springer Nature, 2022, pp. 350–365.","mla":"Prach, Bernd, and Christoph Lampert. “Almost-Orthogonal Layers for Efficient General-Purpose Lipschitz Networks.” Computer Vision – ECCV 2022, vol. 13681, Springer Nature, 2022, pp. 350–65, doi:10.1007/978-3-031-19803-8_21.","chicago":"Prach, Bernd, and Christoph Lampert. “Almost-Orthogonal Layers for Efficient General-Purpose Lipschitz Networks.” In Computer Vision – ECCV 2022, 13681:350–65. Springer Nature, 2022. https://doi.org/10.1007/978-3-031-19803-8_21.","ama":"Prach B, Lampert C. Almost-orthogonal layers for efficient general-purpose Lipschitz networks. In: Computer Vision – ECCV 2022. Vol 13681. Springer Nature; 2022:350-365. doi:10.1007/978-3-031-19803-8_21","apa":"Prach, B., & Lampert, C. (2022). Almost-orthogonal layers for efficient general-purpose Lipschitz networks. In Computer Vision – ECCV 2022 (Vol. 13681, pp. 350–365). Tel Aviv, Israel: Springer Nature. https://doi.org/10.1007/978-3-031-19803-8_21","ieee":"B. Prach and C. Lampert, “Almost-orthogonal layers for efficient general-purpose Lipschitz networks,” in Computer Vision – ECCV 2022, Tel Aviv, Israel, 2022, vol. 13681, pp. 350–365.","ista":"Prach B, Lampert C. 2022. Almost-orthogonal layers for efficient general-purpose Lipschitz networks. Computer Vision – ECCV 2022. ECCV: European Conference on Computer Vision, LNCS, vol. 13681, 350–365."},"publication":"Computer Vision – ECCV 2022","date_published":"2022-10-23T00:00:00Z","scopus_import":"1","article_processing_charge":"No","day":"23"},{"article_processing_charge":"No","publication_identifier":{"isbn":["9781665439022"]},"month":"01","day":"13","citation":{"ama":"Lampert J, Lampert C. Overcoming rare-language discrimination in multi-lingual sentiment analysis. In: 2021 IEEE International Conference on Big Data. IEEE; 2022:5185-5192. doi:10.1109/bigdata52589.2021.9672003","ieee":"J. Lampert and C. Lampert, “Overcoming rare-language discrimination in multi-lingual sentiment analysis,” in 2021 IEEE International Conference on Big Data, Orlando, FL, United States, 2022, pp. 5185–5192.","apa":"Lampert, J., & Lampert, C. (2022). Overcoming rare-language discrimination in multi-lingual sentiment analysis. In 2021 IEEE International Conference on Big Data (pp. 5185–5192). Orlando, FL, United States: IEEE. https://doi.org/10.1109/bigdata52589.2021.9672003","ista":"Lampert J, Lampert C. 2022. Overcoming rare-language discrimination in multi-lingual sentiment analysis. 2021 IEEE International Conference on Big Data. Big Data: International Conference on Big Data, 5185–5192.","short":"J. Lampert, C. Lampert, in:, 2021 IEEE International Conference on Big Data, IEEE, 2022, pp. 5185–5192.","mla":"Lampert, Jasmin, and Christoph Lampert. “Overcoming Rare-Language Discrimination in Multi-Lingual Sentiment Analysis.” 2021 IEEE International Conference on Big Data, IEEE, 2022, pp. 5185–92, doi:10.1109/bigdata52589.2021.9672003.","chicago":"Lampert, Jasmin, and Christoph Lampert. “Overcoming Rare-Language Discrimination in Multi-Lingual Sentiment Analysis.” In 2021 IEEE International Conference on Big Data, 5185–92. IEEE, 2022. https://doi.org/10.1109/bigdata52589.2021.9672003."},"external_id":{"isi":["000800559505036"]},"publication":"2021 IEEE International Conference on Big Data","page":"5185-5192","quality_controlled":"1","isi":1,"date_published":"2022-01-13T00:00:00Z","doi":"10.1109/bigdata52589.2021.9672003","conference":{"location":"Orlando, FL, United States","start_date":"2021-12-15","end_date":"2021-12-18","name":"Big Data: International Conference on Big Data"},"language":[{"iso":"eng"}],"type":"conference","abstract":[{"text":"The digitalization of almost all aspects of our everyday lives has led to unprecedented amounts of data being freely available on the Internet. In particular social media platforms provide rich sources of user-generated data, though typically in unstructured form, and with high diversity, such as written in many different languages. Automatically identifying meaningful information in such big data resources and extracting it efficiently is one of the ongoing challenges of our time. A common step for this is sentiment analysis, which forms the foundation for tasks such as opinion mining or trend prediction. Unfortunately, publicly available tools for this task are almost exclusively available for English-language texts. Consequently, a large fraction of the Internet users, who do not communicate in English, are ignored in automatized studies, a phenomenon called rare-language discrimination.In this work we propose a technique to overcome this problem by a truly multi-lingual model, which can be trained automatically without linguistic knowledge or even the ability to read the many target languages. The main step is to combine self-annotation, specifically the use of emoticons as a proxy for labels, with multi-lingual sentence representations.To evaluate our method we curated several large datasets from data obtained via the free Twitter streaming API. The results show that our proposed multi-lingual training is able to achieve sentiment predictions at the same quality level for rare languages as for frequent ones, and in particular clearly better than what mono-lingual training achieves on the same data. ","lang":"eng"}],"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","_id":"10752","year":"2022","publisher":"IEEE","department":[{"_id":"ChLa"}],"status":"public","title":"Overcoming rare-language discrimination in multi-lingual sentiment analysis","publication_status":"published","author":[{"last_name":"Lampert","first_name":"Jasmin","full_name":"Lampert, Jasmin"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-4561-241X","first_name":"Christoph","last_name":"Lampert"}],"oa_version":"None","date_created":"2022-02-10T14:08:23Z","date_updated":"2023-08-02T14:27:50Z"},{"status":"public","title":"Lightweight conditional model extrapolation for streaming data under class-prior shift","intvolume":" 2022","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","_id":"12161","oa_version":"Preprint","type":"conference","abstract":[{"text":"We introduce LIMES, a new method for learning with non-stationary streaming data, inspired by the recent success of meta-learning. The main idea is not to attempt to learn a single classifier that would have to work well across all occurring data distributions, nor many separate classifiers, but to exploit a hybrid strategy: we learn a single set of model parameters from which a specific classifier for any specific data distribution is derived via classifier adaptation. Assuming a multiclass classification setting with class-prior shift, the adaptation step can be performed analytically with only the classifier’s bias terms being affected. Another contribution of our work is an extrapolation step that predicts suitable adaptation parameters for future time steps based on the previous data. In combination, we obtain a lightweight procedure for learning from streaming data with varying class distribution that adds no trainable parameters and almost no memory or computational overhead compared to training a single model. Experiments on a set of exemplary tasks using Twitter data show that LIMES achieves higher accuracy than alternative approaches, especially with respect to the relevant real-world metric of lowest within-day accuracy.","lang":"eng"}],"page":"2128-2134","publication":"26th International Conference on Pattern Recognition","citation":{"chicago":"Tomaszewska, Paulina, and Christoph Lampert. “Lightweight Conditional Model Extrapolation for Streaming Data under Class-Prior Shift.” In 26th International Conference on Pattern Recognition, 2022:2128–34. Institute of Electrical and Electronics Engineers, 2022. https://doi.org/10.1109/icpr56361.2022.9956195.","mla":"Tomaszewska, Paulina, and Christoph Lampert. “Lightweight Conditional Model Extrapolation for Streaming Data under Class-Prior Shift.” 26th International Conference on Pattern Recognition, vol. 2022, Institute of Electrical and Electronics Engineers, 2022, pp. 2128–34, doi:10.1109/icpr56361.2022.9956195.","short":"P. Tomaszewska, C. Lampert, in:, 26th International Conference on Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 2128–2134.","ista":"Tomaszewska P, Lampert C. 2022. Lightweight conditional model extrapolation for streaming data under class-prior shift. 26th International Conference on Pattern Recognition. ICPR: International Conference on Pattern Recognition vol. 2022, 2128–2134.","ieee":"P. Tomaszewska and C. Lampert, “Lightweight conditional model extrapolation for streaming data under class-prior shift,” in 26th International Conference on Pattern Recognition, Montreal, Canada, 2022, vol. 2022, pp. 2128–2134.","apa":"Tomaszewska, P., & Lampert, C. (2022). Lightweight conditional model extrapolation for streaming data under class-prior shift. In 26th International Conference on Pattern Recognition (Vol. 2022, pp. 2128–2134). Montreal, Canada: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/icpr56361.2022.9956195","ama":"Tomaszewska P, Lampert C. Lightweight conditional model extrapolation for streaming data under class-prior shift. In: 26th International Conference on Pattern Recognition. Vol 2022. Institute of Electrical and Electronics Engineers; 2022:2128-2134. doi:10.1109/icpr56361.2022.9956195"},"date_published":"2022-11-29T00:00:00Z","scopus_import":"1","day":"29","article_processing_charge":"No","publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"Institute of Electrical and Electronics Engineers","year":"2022","date_updated":"2023-08-04T09:06:34Z","date_created":"2023-01-12T12:09:38Z","volume":2022,"author":[{"full_name":"Tomaszewska, Paulina","last_name":"Tomaszewska","first_name":"Paulina"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph"}],"isi":1,"quality_controlled":"1","oa":1,"external_id":{"arxiv":["2206.05181"],"isi":["000897707602018"]},"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2206.05181","open_access":"1"}],"language":[{"iso":"eng"}],"conference":{"end_date":"2022-08-25","location":"Montreal, Canada","start_date":"2022-08-21","name":"ICPR: International Conference on Pattern Recognition"},"doi":"10.1109/icpr56361.2022.9956195","month":"11","publication_identifier":{"eissn":["2831-7475"],"eisbn":["9781665490627"]}},{"oa_version":"Preprint","status":"public","title":"How well do sparse ImageNet models transfer?","_id":"12299","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","abstract":[{"lang":"eng","text":"Transfer learning is a classic paradigm by which models pretrained on large “upstream” datasets are adapted to yield good results on “downstream” specialized datasets. Generally, more accurate models on the “upstream” dataset tend to provide better transfer accuracy “downstream”. In this work, we perform an in-depth investigation of this phenomenon in the context of convolutional neural networks (CNNs) trained on the ImageNet dataset, which have been pruned-that is, compressed by sparsifiying their connections. We consider transfer using unstructured pruned models obtained by applying several state-of-the-art pruning methods, including magnitude-based, second-order, regrowth, lottery-ticket, and regularization approaches, in the context of twelve standard transfer tasks. In a nutshell, our study shows that sparse models can match or even outperform the transfer performance of dense models, even at high sparsities, and, while doing so, can lead to significant inference and even training speedups. At the same time, we observe and analyze significant differences in the behaviour of different pruning methods. The code is available at: https://github.com/IST-DASLab/sparse-imagenet-transfer."}],"type":"conference","date_published":"2022-09-27T00:00:00Z","page":"12256-12266","citation":{"mla":"Iofinova, Eugenia B., et al. “How Well Do Sparse ImageNet Models Transfer?” 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 12256–66, doi:10.1109/cvpr52688.2022.01195.","short":"E.B. Iofinova, E.-A. Peste, M. Kurtz, D.-A. Alistarh, in:, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Institute of Electrical and Electronics Engineers, 2022, pp. 12256–12266.","chicago":"Iofinova, Eugenia B, Elena-Alexandra Peste, Mark Kurtz, and Dan-Adrian Alistarh. “How Well Do Sparse ImageNet Models Transfer?” In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12256–66. Institute of Electrical and Electronics Engineers, 2022. https://doi.org/10.1109/cvpr52688.2022.01195.","ama":"Iofinova EB, Peste E-A, Kurtz M, Alistarh D-A. How well do sparse ImageNet models transfer? In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Institute of Electrical and Electronics Engineers; 2022:12256-12266. doi:10.1109/cvpr52688.2022.01195","ista":"Iofinova EB, Peste E-A, Kurtz M, Alistarh D-A. 2022. How well do sparse ImageNet models transfer? 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Computer Vision and Pattern Recognition, 12256–12266.","ieee":"E. B. Iofinova, E.-A. Peste, M. Kurtz, and D.-A. Alistarh, “How well do sparse ImageNet models transfer?,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, United States, 2022, pp. 12256–12266.","apa":"Iofinova, E. B., Peste, E.-A., Kurtz, M., & Alistarh, D.-A. (2022). How well do sparse ImageNet models transfer? In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 12256–12266). New Orleans, LA, United States: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/cvpr52688.2022.01195"},"publication":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition","article_processing_charge":"No","day":"27","scopus_import":"1","date_created":"2023-01-16T10:06:00Z","date_updated":"2023-08-04T10:33:28Z","related_material":{"record":[{"id":"13074","relation":"dissertation_contains","status":"public"}]},"author":[{"full_name":"Iofinova, Eugenia B","first_name":"Eugenia B","last_name":"Iofinova","id":"f9a17499-f6e0-11ea-865d-fdf9a3f77117","orcid":"0000-0002-7778-3221"},{"full_name":"Peste, Elena-Alexandra","last_name":"Peste","first_name":"Elena-Alexandra","id":"32D78294-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Kurtz","first_name":"Mark","full_name":"Kurtz, Mark"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0003-3650-940X","first_name":"Dan-Adrian","last_name":"Alistarh","full_name":"Alistarh, Dan-Adrian"}],"publisher":"Institute of Electrical and Electronics Engineers","department":[{"_id":"DaAl"},{"_id":"ChLa"}],"publication_status":"published","acknowledgement":"he authors would like to sincerely thank Christoph Lampert and Nir Shavit for fruitful discussions during the development of this work, and Eldar Kurtic for experimental support. EI was supported in part by the FWF DK VGSCO, grant agreement number W1260-N35, while AP and DA acknowledge generous support by the ERC, via Starting Grant 805223 ScaleML.","year":"2022","ec_funded":1,"language":[{"iso":"eng"}],"doi":"10.1109/cvpr52688.2022.01195","conference":{"location":"New Orleans, LA, United States","start_date":"2022-06-18","end_date":"2022-06-24","name":"CVPR: Computer Vision and Pattern Recognition"},"project":[{"name":"Vienna Graduate School on Computational Optimization","_id":"9B9290DE-BA93-11EA-9121-9846C619BF3A","grant_number":" W1260-N35"},{"name":"Elastic Coordination for Scalable Machine Learning","call_identifier":"H2020","_id":"268A44D6-B435-11E9-9278-68D0E5697425","grant_number":"805223"}],"isi":1,"quality_controlled":"1","external_id":{"arxiv":["2111.13445"],"isi":["000870759105034"]},"oa":1,"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2111.13445","open_access":"1"}],"publication_identifier":{"eissn":["2575-7075"]},"month":"09"},{"type":"journal_article","abstract":[{"text":"Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. While many approaches have been developed for training fair models from data, little is known about the robustness of these methods to data corruption. In this work we consider fairness-aware learning under worst-case data manipulations. We show that an adversary can in some situations force any learner to return an overly biased classifier, regardless of the sample size and with or without degrading\r\naccuracy, and that the strength of the excess bias increases for learning problems with underrepresented protected groups in the data. We also prove that our hardness results are tight up to constant factors. To this end, we study two natural learning algorithms that optimize for both accuracy and fairness and show that these algorithms enjoy guarantees that are order-optimal in terms of the corruption ratio and the protected groups frequencies in the large data\r\nlimit.","lang":"eng"}],"ddc":["004"],"title":"Fairness-aware PAC learning from corrupted data","status":"public","intvolume":" 23","_id":"10802","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","file":[{"checksum":"9cac897b54a0ddf3a553a2c33e88cfda","success":1,"date_created":"2022-07-12T15:08:28Z","date_updated":"2022-07-12T15:08:28Z","relation":"main_file","file_id":"11570","file_size":551862,"content_type":"application/pdf","creator":"kschuh","access_level":"open_access","file_name":"2022_JournalMachineLearningResearch_Konstantinov.pdf"}],"oa_version":"Published Version","keyword":["Fairness","robustness","data poisoning","trustworthy machine learning","PAC learning"],"scopus_import":"1","day":"01","has_accepted_license":"1","article_processing_charge":"No","article_type":"original","page":"1-60","publication":"Journal of Machine Learning Research","citation":{"ieee":"N. H. Konstantinov and C. Lampert, “Fairness-aware PAC learning from corrupted data,” Journal of Machine Learning Research, vol. 23. ML Research Press, pp. 1–60, 2022.","apa":"Konstantinov, N. H., & Lampert, C. (2022). Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. ML Research Press.","ista":"Konstantinov NH, Lampert C. 2022. Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. 23, 1–60.","ama":"Konstantinov NH, Lampert C. Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. 2022;23:1-60.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “Fairness-Aware PAC Learning from Corrupted Data.” Journal of Machine Learning Research. ML Research Press, 2022.","short":"N.H. Konstantinov, C. Lampert, Journal of Machine Learning Research 23 (2022) 1–60.","mla":"Konstantinov, Nikola H., and Christoph Lampert. “Fairness-Aware PAC Learning from Corrupted Data.” Journal of Machine Learning Research, vol. 23, ML Research Press, 2022, pp. 1–60."},"date_published":"2022-05-01T00:00:00Z","file_date_updated":"2022-07-12T15:08:28Z","publication_status":"published","publisher":"ML Research Press","department":[{"_id":"ChLa"}],"year":"2022","acknowledgement":"The authors thank Eugenia Iofinova and Bernd Prach for providing feedback on early versions of this paper. This publication was made possible by an ETH AI Center postdoctoral fellowship to Nikola Konstantinov.","date_updated":"2023-09-26T10:44:37Z","date_created":"2022-02-28T14:05:42Z","volume":23,"author":[{"full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","first_name":"Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-4561-241X","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"}],"related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"10799"},{"relation":"shorter_version","status":"public","id":"13241"}]},"month":"05","publication_identifier":{"issn":["1532-4435"],"eissn":["1533-7928"]},"quality_controlled":"1","external_id":{"arxiv":["2102.06004"]},"tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"oa":1,"language":[{"iso":"eng"}]},{"abstract":[{"text":"Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. Many approaches for training fair models from data have been developed and an implicit assumption about such algorithms is that they are able to recover a fair model, despite potential historical biases in the data. In this work we show a number of impossibility results that indicate that there is no learning algorithm that can recover a fair model when a proportion of the dataset is subject to arbitrary manipulations. Specifically, we prove that there are situations in which an adversary can force any learner to return a biased classifier, with or without degrading accuracy, and that the strength of this bias increases for learning problems with underrepresented protected groups in the data. Our results emphasize on the importance of studying further data corruption models of various strength and of establishing stricter data collection practices for fairness-aware learning.","lang":"eng"}],"type":"conference","oa_version":"Preprint","intvolume":" 171","status":"public","title":"On the impossibility of fairness-aware learning from corrupted data","_id":"13241","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","article_processing_charge":"No","day":"01","scopus_import":"1","date_published":"2022-12-01T00:00:00Z","page":"59-83","citation":{"ama":"Konstantinov NH, Lampert C. On the impossibility of fairness-aware learning from corrupted data. In: Proceedings of Machine Learning Research. Vol 171. ML Research Press; 2022:59-83.","ista":"Konstantinov NH, Lampert C. 2022. On the impossibility of fairness-aware learning from corrupted data. Proceedings of Machine Learning Research. vol. 171, 59–83.","apa":"Konstantinov, N. H., & Lampert, C. (2022). On the impossibility of fairness-aware learning from corrupted data. In Proceedings of Machine Learning Research (Vol. 171, pp. 59–83). ML Research Press.","ieee":"N. H. Konstantinov and C. Lampert, “On the impossibility of fairness-aware learning from corrupted data,” in Proceedings of Machine Learning Research, 2022, vol. 171, pp. 59–83.","mla":"Konstantinov, Nikola H., and Christoph Lampert. “On the Impossibility of Fairness-Aware Learning from Corrupted Data.” Proceedings of Machine Learning Research, vol. 171, ML Research Press, 2022, pp. 59–83.","short":"N.H. Konstantinov, C. Lampert, in:, Proceedings of Machine Learning Research, ML Research Press, 2022, pp. 59–83.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “On the Impossibility of Fairness-Aware Learning from Corrupted Data.” In Proceedings of Machine Learning Research, 171:59–83. ML Research Press, 2022."},"publication":"Proceedings of Machine Learning Research","volume":171,"date_updated":"2023-09-26T10:44:37Z","date_created":"2023-07-16T22:01:13Z","related_material":{"record":[{"id":"10802","status":"public","relation":"extended_version"}]},"author":[{"full_name":"Konstantinov, Nikola H","last_name":"Konstantinov","first_name":"Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph"}],"department":[{"_id":"ChLa"}],"publisher":"ML Research Press","publication_status":"published","year":"2022","acknowledgement":"This paper is a shortened, workshop version of Konstantinov and Lampert (2021),\r\nhttps://arxiv.org/abs/2102.06004. For further results, including an analysis of algorithms achieving the lower bounds from this paper, we refer to the full version.","publication_identifier":{"eissn":["2640-3498"]},"month":"12","language":[{"iso":"eng"}],"quality_controlled":"1","external_id":{"arxiv":["2102.06004"]},"oa":1,"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2102.06004"}]},{"month":"03","publication_identifier":{"issn":["2663-337X"],"isbn":["978-3-99078-015-2"]},"project":[{"_id":"2564DBCA-B435-11E9-9278-68D0E5697425","grant_number":"665385","name":"International IST Doctoral Program","call_identifier":"H2020"}],"oa":1,"supervisor":[{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"degree_awarded":"PhD","language":[{"iso":"eng"}],"doi":"10.15479/at:ista:10799","file_date_updated":"2022-03-10T12:11:48Z","ec_funded":1,"publication_status":"published","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"publisher":"Institute of Science and Technology Austria","year":"2022","date_updated":"2023-10-17T12:31:54Z","date_created":"2022-02-28T13:03:49Z","author":[{"id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","last_name":"Konstantinov","first_name":"Nikola H","full_name":"Konstantinov, Nikola H"}],"related_material":{"record":[{"id":"8724","status":"public","relation":"part_of_dissertation"},{"status":"public","relation":"part_of_dissertation","id":"10803"},{"id":"10802","relation":"part_of_dissertation","status":"public"},{"id":"6590","relation":"part_of_dissertation","status":"public"}]},"keyword":["robustness","fairness","machine learning","PAC learning","adversarial learning"],"day":"08","article_processing_charge":"No","has_accepted_license":"1","page":"176","citation":{"chicago":"Konstantinov, Nikola H. “Robustness and Fairness in Machine Learning.” Institute of Science and Technology Austria, 2022. https://doi.org/10.15479/at:ista:10799.","mla":"Konstantinov, Nikola H. Robustness and Fairness in Machine Learning. Institute of Science and Technology Austria, 2022, doi:10.15479/at:ista:10799.","short":"N.H. Konstantinov, Robustness and Fairness in Machine Learning, Institute of Science and Technology Austria, 2022.","ista":"Konstantinov NH. 2022. Robustness and fairness in machine learning. Institute of Science and Technology Austria.","apa":"Konstantinov, N. H. (2022). Robustness and fairness in machine learning. Institute of Science and Technology Austria. https://doi.org/10.15479/at:ista:10799","ieee":"N. H. Konstantinov, “Robustness and fairness in machine learning,” Institute of Science and Technology Austria, 2022.","ama":"Konstantinov NH. Robustness and fairness in machine learning. 2022. doi:10.15479/at:ista:10799"},"date_published":"2022-03-08T00:00:00Z","alternative_title":["ISTA Thesis"],"type":"dissertation","abstract":[{"lang":"eng","text":"Because of the increasing popularity of machine learning methods, it is becoming important to understand the impact of learned components on automated decision-making systems and to guarantee that their consequences are beneficial to society. In other words, it is necessary to ensure that machine learning is sufficiently trustworthy to be used in real-world applications. This thesis studies two properties of machine learning models that are highly desirable for the\r\nsake of reliability: robustness and fairness. In the first part of the thesis we study the robustness of learning algorithms to training data corruption. Previous work has shown that machine learning models are vulnerable to a range\r\nof training set issues, varying from label noise through systematic biases to worst-case data manipulations. This is an especially relevant problem from a present perspective, since modern machine learning methods are particularly data hungry and therefore practitioners often have to rely on data collected from various external sources, e.g. from the Internet, from app users or via crowdsourcing. Naturally, such sources vary greatly in the quality and reliability of the\r\ndata they provide. With these considerations in mind, we study the problem of designing machine learning algorithms that are robust to corruptions in data coming from multiple sources. We show that, in contrast to the case of a single dataset with outliers, successful learning within this model is possible both theoretically and practically, even under worst-case data corruptions. The second part of this thesis deals with fairness-aware machine learning. There are multiple areas where machine learning models have shown promising results, but where careful considerations are required, in order to avoid discrimanative decisions taken by such learned components. Ensuring fairness can be particularly challenging, because real-world training datasets are expected to contain various forms of historical bias that may affect the learning process. In this thesis we show that data corruption can indeed render the problem of achieving fairness impossible, by tightly characterizing the theoretical limits of fair learning under worst-case data manipulations. However, assuming access to clean data, we also show how fairness-aware learning can be made practical in contexts beyond binary classification, in particular in the challenging learning to rank setting."}],"ddc":["000"],"title":"Robustness and fairness in machine learning","status":"public","_id":"10799","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","file":[{"relation":"main_file","file_id":"10823","date_updated":"2022-03-06T11:42:54Z","date_created":"2022-03-06T11:42:54Z","checksum":"626bc523ae8822d20e635d0e2d95182e","success":1,"file_name":"thesis.pdf","access_level":"open_access","file_size":4204905,"content_type":"application/pdf","creator":"nkonstan"},{"file_id":"10824","relation":"source_file","checksum":"e2ca2b88350ac8ea1515b948885cbcb1","date_created":"2022-03-06T11:42:57Z","date_updated":"2022-03-10T12:11:48Z","access_level":"closed","file_name":"thesis.zip","creator":"nkonstan","content_type":"application/x-zip-compressed","file_size":22841103}],"oa_version":"Published Version"},{"language":[{"iso":"eng"}],"conference":{"name":"DAGM GCPR: German Conference on Pattern Recognition ","end_date":"2020-10-01","location":"Tübingen, Germany","start_date":"2020-09-28"},"doi":"10.1007/978-3-030-71278-5_18","quality_controlled":"1","oa":1,"month":"03","publication_identifier":{"issn":["0302-9743"],"isbn":["9783030712778"],"eissn":["1611-3349"]},"date_updated":"2022-08-12T07:28:47Z","date_created":"2021-03-01T09:01:16Z","volume":12544,"author":[{"full_name":"Volhejn, Vaclav","last_name":"Volhejn","first_name":"Vaclav","id":"d5235fb4-7a6d-11eb-b254-f25d12d631a8"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"publication_status":"published","publisher":"Springer","department":[{"_id":"ChLa"}],"year":"2021","file_date_updated":"2022-08-12T07:27:58Z","date_published":"2021-03-17T00:00:00Z","page":"246-259","publication":"42nd German Conference on Pattern Recognition","citation":{"ieee":"V. Volhejn and C. Lampert, “Does SGD implicitly optimize for smoothness?,” in 42nd German Conference on Pattern Recognition, Tübingen, Germany, 2021, vol. 12544, pp. 246–259.","apa":"Volhejn, V., & Lampert, C. (2021). Does SGD implicitly optimize for smoothness? In 42nd German Conference on Pattern Recognition (Vol. 12544, pp. 246–259). Tübingen, Germany: Springer. https://doi.org/10.1007/978-3-030-71278-5_18","ista":"Volhejn V, Lampert C. 2021. Does SGD implicitly optimize for smoothness? 42nd German Conference on Pattern Recognition. DAGM GCPR: German Conference on Pattern Recognition LNCS vol. 12544, 246–259.","ama":"Volhejn V, Lampert C. Does SGD implicitly optimize for smoothness? In: 42nd German Conference on Pattern Recognition. Vol 12544. LNCS. Springer; 2021:246-259. doi:10.1007/978-3-030-71278-5_18","chicago":"Volhejn, Vaclav, and Christoph Lampert. “Does SGD Implicitly Optimize for Smoothness?” In 42nd German Conference on Pattern Recognition, 12544:246–59. LNCS. Springer, 2021. https://doi.org/10.1007/978-3-030-71278-5_18.","short":"V. Volhejn, C. Lampert, in:, 42nd German Conference on Pattern Recognition, Springer, 2021, pp. 246–259.","mla":"Volhejn, Vaclav, and Christoph Lampert. “Does SGD Implicitly Optimize for Smoothness?” 42nd German Conference on Pattern Recognition, vol. 12544, Springer, 2021, pp. 246–59, doi:10.1007/978-3-030-71278-5_18."},"day":"17","has_accepted_license":"1","article_processing_charge":"No","series_title":"LNCS","scopus_import":"1","file":[{"creator":"dernst","content_type":"application/pdf","file_size":420234,"access_level":"open_access","file_name":"2020_GCPR_submitted_Volhejn.pdf","success":1,"checksum":"3e3628ab1cf658d82524963f808004ea","date_updated":"2022-08-12T07:27:58Z","date_created":"2022-08-12T07:27:58Z","file_id":"11820","relation":"main_file"}],"oa_version":"Submitted Version","status":"public","ddc":["510"],"title":"Does SGD implicitly optimize for smoothness?","intvolume":" 12544","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"9210","abstract":[{"text":"Modern neural networks can easily fit their training set perfectly. Surprisingly, despite being “overfit” in this way, they tend to generalize well to future data, thereby defying the classic bias–variance trade-off of machine learning theory. Of the many possible explanations, a prevalent one is that training by stochastic gradient descent (SGD) imposes an implicit bias that leads it to learn simple functions, and these simple functions generalize well. However, the specifics of this implicit bias are not well understood.\r\nIn this work, we explore the smoothness conjecture which states that SGD is implicitly biased towards learning functions that are smooth. We propose several measures to formalize the intuitive notion of smoothness, and we conduct experiments to determine whether SGD indeed implicitly optimizes for these measures. Our findings rule out the possibility that smoothness measures based on first-order derivatives are being implicitly enforced. They are supportive, though, of the smoothness conjecture for measures based on second-order derivatives.","lang":"eng"}],"type":"conference"},{"quality_controlled":"1","citation":{"ama":"Phuong M, Lampert C. The inductive bias of ReLU networks on orthogonally separable data. In: 9th International Conference on Learning Representations. ; 2021.","apa":"Phuong, M., & Lampert, C. (2021). The inductive bias of ReLU networks on orthogonally separable data. In 9th International Conference on Learning Representations. Virtual.","ieee":"M. Phuong and C. Lampert, “The inductive bias of ReLU networks on orthogonally separable data,” in 9th International Conference on Learning Representations, Virtual, 2021.","ista":"Phuong M, Lampert C. 2021. The inductive bias of ReLU networks on orthogonally separable data. 9th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","short":"M. Phuong, C. Lampert, in:, 9th International Conference on Learning Representations, 2021.","mla":"Phuong, Mary, and Christoph Lampert. “The Inductive Bias of ReLU Networks on Orthogonally Separable Data.” 9th International Conference on Learning Representations, 2021.","chicago":"Phuong, Mary, and Christoph Lampert. “The Inductive Bias of ReLU Networks on Orthogonally Separable Data.” In 9th International Conference on Learning Representations, 2021."},"main_file_link":[{"url":"https://openreview.net/pdf?id=krz7T0xU9Z_","open_access":"1"}],"oa":1,"publication":"9th International Conference on Learning Representations","language":[{"iso":"eng"}],"date_published":"2021-05-01T00:00:00Z","conference":{"name":" ICLR: International Conference on Learning Representations","location":"Virtual","start_date":"2021-05-03","end_date":"2021-05-07"},"scopus_import":"1","has_accepted_license":"1","article_processing_charge":"No","month":"05","day":"01","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"ddc":["000"],"title":"The inductive bias of ReLU networks on orthogonally separable data","publication_status":"published","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"9416","year":"2021","oa_version":"Published Version","file":[{"relation":"main_file","file_id":"9417","date_updated":"2021-05-24T11:15:57Z","date_created":"2021-05-24T11:15:57Z","checksum":"f34ff17017527db5ba6927f817bdd125","file_name":"iclr2021_conference.pdf","access_level":"open_access","file_size":502356,"content_type":"application/pdf","creator":"bphuong"}],"date_created":"2021-05-24T11:16:46Z","date_updated":"2023-09-07T13:29:50Z","related_material":{"record":[{"id":"9418","relation":"dissertation_contains","status":"public"}]},"author":[{"full_name":"Bui Thi Mai, Phuong","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","last_name":"Bui Thi Mai","first_name":"Phuong"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"type":"conference","file_date_updated":"2021-05-24T11:15:57Z","abstract":[{"text":"We study the inductive bias of two-layer ReLU networks trained by gradient flow. We identify a class of easy-to-learn (`orthogonally separable') datasets, and characterise the solution that ReLU networks trained on such datasets converge to. Irrespective of network width, the solution turns out to be a combination of two max-margin classifiers: one corresponding to the positive data subset and one corresponding to the negative data subset. The proof is based on the recently introduced concept of extremal sectors, for which we prove a number of properties in the context of orthogonal separability. In particular, we prove stationarity of activation patterns from some time onwards, which enables a reduction of the ReLU network to an ensemble of linear subnetworks.","lang":"eng"}]},{"article_processing_charge":"No","day":"07","month":"06","citation":{"short":"N.H. Konstantinov, C. Lampert, ArXiv (n.d.).","mla":"Konstantinov, Nikola H., and Christoph Lampert. “Fairness through Regularization for Learning to Rank.” ArXiv, 2102.05996, doi:10.48550/arXiv.2102.05996.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “Fairness through Regularization for Learning to Rank.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2102.05996.","ama":"Konstantinov NH, Lampert C. Fairness through regularization for learning to rank. arXiv. doi:10.48550/arXiv.2102.05996","ieee":"N. H. Konstantinov and C. Lampert, “Fairness through regularization for learning to rank,” arXiv. .","apa":"Konstantinov, N. H., & Lampert, C. (n.d.). Fairness through regularization for learning to rank. arXiv. https://doi.org/10.48550/arXiv.2102.05996","ista":"Konstantinov NH, Lampert C. Fairness through regularization for learning to rank. arXiv, 2102.05996."},"oa":1,"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2102.05996"}],"external_id":{"arxiv":["2102.05996"]},"publication":"arXiv","date_published":"2021-06-07T00:00:00Z","doi":"10.48550/arXiv.2102.05996","language":[{"iso":"eng"}],"type":"preprint","article_number":"2102.05996","abstract":[{"lang":"eng","text":"Given the abundance of applications of ranking in recent years, addressing fairness concerns around automated ranking systems becomes necessary for increasing the trust among end-users. Previous work on fair ranking has mostly focused on application-specific fairness notions, often tailored to online advertising, and it rarely considers learning as part of the process. In this work, we show how to transfer numerous fairness notions from binary classification to a learning to rank setting. Our formalism allows us to design methods for incorporating fairness objectives with provable generalization guarantees. An extensive experimental evaluation shows that our method can improve ranking fairness substantially with no or only little loss of model quality."}],"year":"2021","_id":"10803","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","department":[{"_id":"ChLa"}],"status":"public","title":"Fairness through regularization for learning to rank","publication_status":"submitted","related_material":{"record":[{"id":"10799","relation":"dissertation_contains","status":"public"}]},"author":[{"full_name":"Konstantinov, Nikola H","first_name":"Nikola H","last_name":"Konstantinov","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-4561-241X"}],"oa_version":"Preprint","date_updated":"2023-09-07T13:42:08Z","date_created":"2022-02-28T14:13:59Z"},{"year":"2021","publisher":"Institute of Science and Technology Austria","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"publication_status":"published","related_material":{"record":[{"relation":"part_of_dissertation","status":"deleted","id":"7435"},{"relation":"part_of_dissertation","status":"public","id":"7481"},{"id":"9416","status":"public","relation":"part_of_dissertation"},{"id":"7479","relation":"part_of_dissertation","status":"public"}]},"author":[{"id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","first_name":"Phuong","last_name":"Bui Thi Mai","full_name":"Bui Thi Mai, Phuong"}],"date_updated":"2023-09-08T11:11:12Z","date_created":"2021-05-24T13:06:23Z","file_date_updated":"2021-05-24T11:56:02Z","oa":1,"doi":"10.15479/AT:ISTA:9418","language":[{"iso":"eng"}],"acknowledged_ssus":[{"_id":"ScienComp"},{"_id":"CampIT"},{"_id":"E-Lib"}],"degree_awarded":"PhD","supervisor":[{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"}],"publication_identifier":{"issn":["2663-337X"]},"month":"05","_id":"9418","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","status":"public","title":"Underspecification in deep learning","ddc":["000"],"file":[{"access_level":"open_access","file_name":"mph-thesis-v519-pdfimages.pdf","creator":"bphuong","content_type":"application/pdf","file_size":2673905,"file_id":"9419","relation":"main_file","success":1,"checksum":"4f0abe64114cfed264f9d36e8d1197e3","date_updated":"2021-05-24T11:22:29Z","date_created":"2021-05-24T11:22:29Z"},{"access_level":"closed","file_name":"thesis.zip","creator":"bphuong","content_type":"application/zip","file_size":92995100,"file_id":"9420","relation":"source_file","checksum":"f5699e876bc770a9b0df8345a77720a2","date_created":"2021-05-24T11:56:02Z","date_updated":"2021-05-24T11:56:02Z"}],"oa_version":"Published Version","type":"dissertation","alternative_title":["ISTA Thesis"],"abstract":[{"text":"Deep learning is best known for its empirical success across a wide range of applications\r\nspanning computer vision, natural language processing and speech. Of equal significance,\r\nthough perhaps less known, are its ramifications for learning theory: deep networks have\r\nbeen observed to perform surprisingly well in the high-capacity regime, aka the overfitting\r\nor underspecified regime. Classically, this regime on the far right of the bias-variance curve\r\nis associated with poor generalisation; however, recent experiments with deep networks\r\nchallenge this view.\r\n\r\nThis thesis is devoted to investigating various aspects of underspecification in deep learning.\r\nFirst, we argue that deep learning models are underspecified on two levels: a) any given\r\ntraining dataset can be fit by many different functions, and b) any given function can be\r\nexpressed by many different parameter configurations. We refer to the second kind of\r\nunderspecification as parameterisation redundancy and we precisely characterise its extent.\r\nSecond, we characterise the implicit criteria (the inductive bias) that guide learning in the\r\nunderspecified regime. Specifically, we consider a nonlinear but tractable classification\r\nsetting, and show that given the choice, neural networks learn classifiers with a large margin.\r\nThird, we consider learning scenarios where the inductive bias is not by itself sufficient to\r\ndeal with underspecification. We then study different ways of ‘tightening the specification’: i)\r\nIn the setting of representation learning with variational autoencoders, we propose a hand-\r\ncrafted regulariser based on mutual information. ii) In the setting of binary classification, we\r\nconsider soft-label (real-valued) supervision. We derive a generalisation bound for linear\r\nnetworks supervised in this way and verify that soft labels facilitate fast learning. Finally, we\r\nexplore an application of soft-label supervision to the training of multi-exit models.","lang":"eng"}],"citation":{"chicago":"Phuong, Mary. “Underspecification in Deep Learning.” Institute of Science and Technology Austria, 2021. https://doi.org/10.15479/AT:ISTA:9418.","mla":"Phuong, Mary. Underspecification in Deep Learning. Institute of Science and Technology Austria, 2021, doi:10.15479/AT:ISTA:9418.","short":"M. Phuong, Underspecification in Deep Learning, Institute of Science and Technology Austria, 2021.","ista":"Phuong M. 2021. Underspecification in deep learning. Institute of Science and Technology Austria.","apa":"Phuong, M. (2021). Underspecification in deep learning. Institute of Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:9418","ieee":"M. Phuong, “Underspecification in deep learning,” Institute of Science and Technology Austria, 2021.","ama":"Phuong M. Underspecification in deep learning. 2021. doi:10.15479/AT:ISTA:9418"},"page":"125","date_published":"2021-05-30T00:00:00Z","has_accepted_license":"1","article_processing_charge":"No","day":"30"},{"oa_version":"None","date_created":"2024-02-14T14:05:32Z","date_updated":"2024-02-19T10:59:04Z","edition":"2","author":[{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"department":[{"_id":"ChLa"}],"publisher":"Springer","editor":[{"full_name":"Ikeuchi, Katsushi","last_name":"Ikeuchi","first_name":"Katsushi"}],"title":"Zero-Shot Learning","publication_status":"published","status":"public","_id":"14987","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2021","abstract":[{"lang":"eng","text":"The goal of zero-shot learning is to construct a classifier that can identify object classes for which no training examples are available. When training data for some of the object classes is available but not for others, the name generalized zero-shot learning is commonly used.\r\nIn a wider sense, the phrase zero-shot is also used to describe other machine learning-based approaches that require no training data from the problem of interest, such as zero-shot action recognition or zero-shot machine translation."}],"place":"Cham","type":"book_chapter","language":[{"iso":"eng"}],"date_published":"2021-10-13T00:00:00Z","doi":"10.1007/978-3-030-63416-2_874","page":"1395-1397","quality_controlled":"1","citation":{"short":"C. Lampert, in:, K. Ikeuchi (Ed.), Computer Vision, 2nd ed., Springer, Cham, 2021, pp. 1395–1397.","mla":"Lampert, Christoph. “Zero-Shot Learning.” Computer Vision, edited by Katsushi Ikeuchi, 2nd ed., Springer, 2021, pp. 1395–97, doi:10.1007/978-3-030-63416-2_874.","chicago":"Lampert, Christoph. “Zero-Shot Learning.” In Computer Vision, edited by Katsushi Ikeuchi, 2nd ed., 1395–97. Cham: Springer, 2021. https://doi.org/10.1007/978-3-030-63416-2_874.","ama":"Lampert C. Zero-Shot Learning. In: Ikeuchi K, ed. Computer Vision. 2nd ed. Cham: Springer; 2021:1395-1397. doi:10.1007/978-3-030-63416-2_874","apa":"Lampert, C. (2021). Zero-Shot Learning. In K. Ikeuchi (Ed.), Computer Vision (2nd ed., pp. 1395–1397). Cham: Springer. https://doi.org/10.1007/978-3-030-63416-2_874","ieee":"C. Lampert, “Zero-Shot Learning,” in Computer Vision, 2nd ed., K. Ikeuchi, Ed. Cham: Springer, 2021, pp. 1395–1397.","ista":"Lampert C. 2021.Zero-Shot Learning. In: Computer Vision. , 1395–1397."},"publication":"Computer Vision","publication_identifier":{"eisbn":["9783030634162"],"isbn":["9783030634155"]},"article_processing_charge":"No","month":"10","day":"13"},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"8063","year":"2020","department":[{"_id":"ChLa"}],"ddc":["004"],"title":"Object-centric image generation with factored depths, locations, and appearances","status":"public","publication_status":"submitted","author":[{"last_name":"Anciukevicius","first_name":"Titas","full_name":"Anciukevicius, Titas"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"},{"full_name":"Henderson, Paul M","first_name":"Paul M","last_name":"Henderson","id":"13C09E74-18D9-11E9-8878-32CFE5697425","orcid":"0000-0002-5198-7445"}],"oa_version":"Preprint","date_created":"2020-06-29T23:55:23Z","date_updated":"2021-01-12T08:16:44Z","type":"preprint","article_number":"2004.00642","abstract":[{"text":"We present a generative model of images that explicitly reasons over the set\r\nof objects they show. Our model learns a structured latent representation that\r\nseparates objects from each other and from the background; unlike prior works,\r\nit explicitly represents the 2D position and depth of each object, as well as\r\nan embedding of its segmentation mask and appearance. The model can be trained\r\nfrom images alone in a purely unsupervised fashion without the need for object\r\nmasks or depth information. Moreover, it always generates complete objects,\r\neven though a significant fraction of training images contain occlusions.\r\nFinally, we show that our model can infer decompositions of novel images into\r\ntheir constituent objects, including accurate prediction of depth ordering and\r\nsegmentation of occluded parts.","lang":"eng"}],"license":"https://creativecommons.org/licenses/by-sa/4.0/","oa":1,"tmp":{"short":"CC BY-SA (4.0)","image":"/images/cc_by_sa.png","name":"Creative Commons Attribution-ShareAlike 4.0 International Public License (CC BY-SA 4.0)","legal_code_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode"},"external_id":{"arxiv":["2004.00642"]},"citation":{"ama":"Anciukevicius T, Lampert C, Henderson PM. Object-centric image generation with factored depths, locations, and appearances. arXiv.","ieee":"T. Anciukevicius, C. Lampert, and P. M. Henderson, “Object-centric image generation with factored depths, locations, and appearances,” arXiv. .","apa":"Anciukevicius, T., Lampert, C., & Henderson, P. M. (n.d.). Object-centric image generation with factored depths, locations, and appearances. arXiv.","ista":"Anciukevicius T, Lampert C, Henderson PM. Object-centric image generation with factored depths, locations, and appearances. arXiv, 2004.00642.","short":"T. Anciukevicius, C. Lampert, P.M. Henderson, ArXiv (n.d.).","mla":"Anciukevicius, Titas, et al. “Object-Centric Image Generation with Factored Depths, Locations, and Appearances.” ArXiv, 2004.00642.","chicago":"Anciukevicius, Titas, Christoph Lampert, and Paul M Henderson. “Object-Centric Image Generation with Factored Depths, Locations, and Appearances.” ArXiv, n.d."},"main_file_link":[{"url":"https://arxiv.org/abs/2004.00642","open_access":"1"}],"publication":"arXiv","date_published":"2020-04-01T00:00:00Z","language":[{"iso":"eng"}],"article_processing_charge":"No","day":"01","month":"04"},{"acknowledgement":"This research was supported by the Scientific Service Units (SSU) of IST Austria through resources\r\nprovided by Scientific Computing (SciComp). PH is employed part-time by Blackford Analysis, but\r\nthey did not support this project in any way.","year":"2020","department":[{"_id":"ChLa"}],"publisher":"Curran Associates","publication_status":"published","author":[{"last_name":"Henderson","first_name":"Paul M","orcid":"0000-0002-5198-7445","id":"13C09E74-18D9-11E9-8878-32CFE5697425","full_name":"Henderson, Paul M"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"volume":33,"date_created":"2020-07-31T16:59:19Z","date_updated":"2023-04-25T09:49:58Z","main_file_link":[{"url":"https://arxiv.org/abs/2007.06705","open_access":"1"}],"oa":1,"external_id":{"arxiv":["2007.06705"]},"quality_controlled":"1","conference":{"end_date":"2020-12-12","location":"Vancouver, Canada","start_date":"2020-12-06","name":"NeurIPS: Neural Information Processing Systems"},"language":[{"iso":"eng"}],"acknowledged_ssus":[{"_id":"ScienComp"}],"publication_identifier":{"isbn":["9781713829546"]},"month":"07","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"8188","intvolume":" 33","title":"Unsupervised object-centric video generation and decomposition in 3D","status":"public","oa_version":"Preprint","type":"conference","abstract":[{"text":"A natural approach to generative modeling of videos is to represent them as a composition of moving objects. Recent works model a set of 2D sprites over a slowly-varying background, but without considering the underlying 3D scene that\r\ngives rise to them. We instead propose to model a video as the view seen while moving through a scene with multiple 3D objects and a 3D background. Our model is trained from monocular videos without any supervision, yet learns to\r\ngenerate coherent 3D scenes containing several moving objects. We conduct detailed experiments on two datasets, going beyond the visual complexity supported by state-of-the-art generative approaches. We evaluate our method on\r\ndepth-prediction and 3D object detection---tasks which cannot be addressed by those earlier works---and show it out-performs them even on 2D instance segmentation and tracking.","lang":"eng"}],"citation":{"ista":"Henderson PM, Lampert C. 2020. Unsupervised object-centric video generation and decomposition in 3D. 34th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems vol. 33, 3106–3117.","apa":"Henderson, P. M., & Lampert, C. (2020). Unsupervised object-centric video generation and decomposition in 3D. In 34th Conference on Neural Information Processing Systems (Vol. 33, pp. 3106–3117). Vancouver, Canada: Curran Associates.","ieee":"P. M. Henderson and C. Lampert, “Unsupervised object-centric video generation and decomposition in 3D,” in 34th Conference on Neural Information Processing Systems, Vancouver, Canada, 2020, vol. 33, pp. 3106–3117.","ama":"Henderson PM, Lampert C. Unsupervised object-centric video generation and decomposition in 3D. In: 34th Conference on Neural Information Processing Systems. Vol 33. Curran Associates; 2020:3106–3117.","chicago":"Henderson, Paul M, and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” In 34th Conference on Neural Information Processing Systems, 33:3106–3117. Curran Associates, 2020.","mla":"Henderson, Paul M., and Christoph Lampert. “Unsupervised Object-Centric Video Generation and Decomposition in 3D.” 34th Conference on Neural Information Processing Systems, vol. 33, Curran Associates, 2020, pp. 3106–3117.","short":"P.M. Henderson, C. Lampert, in:, 34th Conference on Neural Information Processing Systems, Curran Associates, 2020, pp. 3106–3117."},"publication":"34th Conference on Neural Information Processing Systems","page":"3106–3117","date_published":"2020-07-07T00:00:00Z","article_processing_charge":"No","day":"07"},{"month":"04","publication_identifier":{"eissn":["1573-1405"],"issn":["0920-5691"]},"oa":1,"tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"external_id":{"isi":["000491042100002"],"arxiv":["1901.06447"]},"isi":1,"quality_controlled":"1","project":[{"name":"IST Austria Open Access Fund","_id":"B67AFEDC-15C9-11EA-A837-991A96BB2854"}],"doi":"10.1007/s11263-019-01219-8","language":[{"iso":"eng"}],"file_date_updated":"2020-07-14T12:47:46Z","year":"2020","acknowledgement":"Open access funding provided by Institute of Science and Technology (IST Austria).","publication_status":"published","publisher":"Springer Nature","department":[{"_id":"ChLa"}],"author":[{"orcid":"0000-0002-5198-7445","id":"13C09E74-18D9-11E9-8878-32CFE5697425","last_name":"Henderson","first_name":"Paul M","full_name":"Henderson, Paul M"},{"last_name":"Ferrari","first_name":"Vittorio","full_name":"Ferrari, Vittorio"}],"date_created":"2019-10-17T13:38:20Z","date_updated":"2023-08-17T14:01:16Z","volume":128,"scopus_import":"1","day":"01","article_processing_charge":"Yes (via OA deal)","has_accepted_license":"1","publication":"International Journal of Computer Vision","citation":{"ama":"Henderson PM, Ferrari V. Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. International Journal of Computer Vision. 2020;128:835-854. doi:10.1007/s11263-019-01219-8","apa":"Henderson, P. M., & Ferrari, V. (2020). Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. International Journal of Computer Vision. Springer Nature. https://doi.org/10.1007/s11263-019-01219-8","ieee":"P. M. Henderson and V. Ferrari, “Learning single-image 3D reconstruction by generative modelling of shape, pose and shading,” International Journal of Computer Vision, vol. 128. Springer Nature, pp. 835–854, 2020.","ista":"Henderson PM, Ferrari V. 2020. Learning single-image 3D reconstruction by generative modelling of shape, pose and shading. International Journal of Computer Vision. 128, 835–854.","short":"P.M. Henderson, V. Ferrari, International Journal of Computer Vision 128 (2020) 835–854.","mla":"Henderson, Paul M., and Vittorio Ferrari. “Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading.” International Journal of Computer Vision, vol. 128, Springer Nature, 2020, pp. 835–54, doi:10.1007/s11263-019-01219-8.","chicago":"Henderson, Paul M, and Vittorio Ferrari. “Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading.” International Journal of Computer Vision. Springer Nature, 2020. https://doi.org/10.1007/s11263-019-01219-8."},"article_type":"original","page":"835-854","date_published":"2020-04-01T00:00:00Z","type":"journal_article","abstract":[{"lang":"eng","text":"We present a unified framework tackling two problems: class-specific 3D reconstruction from a single image, and generation of new 3D shape samples. These tasks have received considerable attention recently; however, most existing approaches rely on 3D supervision, annotation of 2D images with keypoints or poses, and/or training with multiple views of each object instance. Our framework is very general: it can be trained in similar settings to existing approaches, while also supporting weaker supervision. Importantly, it can be trained purely from 2D images, without pose annotations, and with only a single view per instance. We employ meshes as an output representation, instead of voxels used in most prior work. This allows us to reason over lighting parameters and exploit shading information during training, which previous 2D-supervised methods cannot. Thus, our method can learn to generate and reconstruct concave object classes. We evaluate our approach in various settings, showing that: (i) it learns to disentangle shape from pose and lighting; (ii) using shading in the loss improves performance compared to just silhouettes; (iii) when using a standard single white light, our model outperforms state-of-the-art 2D-supervised methods, both with and without pose supervision, thanks to exploiting shading cues; (iv) performance improves further when using multiple coloured lights, even approaching that of state-of-the-art 3D-supervised methods; (v) shapes produced by our model capture smooth surfaces and fine details better than voxel-based approaches; and (vi) our approach supports concave classes such as bathtubs and sofas, which methods based on silhouettes cannot learn."}],"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","_id":"6952","ddc":["004"],"status":"public","title":"Learning single-image 3D reconstruction by generative modelling of shape, pose and shading","intvolume":" 128","file":[{"creator":"dernst","file_size":2243134,"content_type":"application/pdf","access_level":"open_access","file_name":"2019_CompVision_Henderson.pdf","checksum":"a0f05dd4f5f64e4f713d8d9d4b5b1e3f","date_updated":"2020-07-14T12:47:46Z","date_created":"2019-10-25T10:28:29Z","file_id":"6973","relation":"main_file"}],"oa_version":"Published Version"},{"article_number":"1716-1725","publication_status":"published","publisher":"IEEE","department":[{"_id":"ChLa"}],"year":"2020","date_updated":"2023-09-07T13:16:17Z","date_created":"2020-06-07T22:00:53Z","author":[{"full_name":"Royer, Amélie","id":"3811D890-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8407-0705","first_name":"Amélie","last_name":"Royer"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"related_material":{"record":[{"id":"8331","status":"deleted","relation":"dissertation_contains"},{"id":"8390","status":"public","relation":"dissertation_contains"}]},"month":"03","publication_identifier":{"isbn":["9781728165530"]},"quality_controlled":"1","oa":1,"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2004.12623"}],"external_id":{"arxiv":["2004.12623"]},"language":[{"iso":"eng"}],"conference":{"location":" Snowmass Village, CO, United States","start_date":"2020-03-01","end_date":"2020-03-05","name":"WACV: Winter Conference on Applications of Computer Vision"},"doi":"10.1109/WACV45572.2020.9093288","type":"conference","abstract":[{"text":"State-of-the-art detection systems are generally evaluated on their ability to exhaustively retrieve objects densely distributed in the image, across a wide variety of appearances and semantic categories. Orthogonal to this, many real-life object detection applications, for example in remote sensing, instead require dealing with large images that contain only a few small objects of a single class, scattered heterogeneously across the space. In addition, they are often subject to strict computational constraints, such as limited battery capacity and computing power.To tackle these more practical scenarios, we propose a novel flexible detection scheme that efficiently adapts to variable object sizes and densities: We rely on a sequence of detection stages, each of which has the ability to predict groups of objects as well as individuals. Similar to a detection cascade, this multi-stage architecture spares computational effort by discarding large irrelevant regions of the image early during the detection process. The ability to group objects provides further computational and memory savings, as it allows working with lower image resolutions in early stages, where groups are more easily detected than individuals, as they are more salient. We report experimental results on two aerial image datasets, and show that the proposed method is as accurate yet computationally more efficient than standard single-shot detectors, consistently across three different backbone architectures.","lang":"eng"}],"title":"Localizing grouped instances for efficient detection in low-resource scenarios","status":"public","_id":"7936","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","scopus_import":1,"day":"01","article_processing_charge":"No","publication":"IEEE Winter Conference on Applications of Computer Vision","citation":{"short":"A. Royer, C. Lampert, in:, IEEE Winter Conference on Applications of Computer Vision, IEEE, 2020.","mla":"Royer, Amélie, and Christoph Lampert. “Localizing Grouped Instances for Efficient Detection in Low-Resource Scenarios.” IEEE Winter Conference on Applications of Computer Vision, 1716–1725, IEEE, 2020, doi:10.1109/WACV45572.2020.9093288.","chicago":"Royer, Amélie, and Christoph Lampert. “Localizing Grouped Instances for Efficient Detection in Low-Resource Scenarios.” In IEEE Winter Conference on Applications of Computer Vision. IEEE, 2020. https://doi.org/10.1109/WACV45572.2020.9093288.","ama":"Royer A, Lampert C. Localizing grouped instances for efficient detection in low-resource scenarios. In: IEEE Winter Conference on Applications of Computer Vision. IEEE; 2020. doi:10.1109/WACV45572.2020.9093288","ieee":"A. Royer and C. Lampert, “Localizing grouped instances for efficient detection in low-resource scenarios,” in IEEE Winter Conference on Applications of Computer Vision, Snowmass Village, CO, United States, 2020.","apa":"Royer, A., & Lampert, C. (2020). Localizing grouped instances for efficient detection in low-resource scenarios. In IEEE Winter Conference on Applications of Computer Vision. Snowmass Village, CO, United States: IEEE. https://doi.org/10.1109/WACV45572.2020.9093288","ista":"Royer A, Lampert C. 2020. Localizing grouped instances for efficient detection in low-resource scenarios. IEEE Winter Conference on Applications of Computer Vision. WACV: Winter Conference on Applications of Computer Vision, 1716–1725."},"date_published":"2020-03-01T00:00:00Z"},{"month":"03","publication_identifier":{"isbn":["9781728165530"]},"conference":{"location":"Snowmass Village, CO, United States","start_date":"2020-03-01","end_date":"2020-03-05","name":"WACV: Winter Conference on Applications of Computer Vision"},"doi":"10.1109/WACV45572.2020.9093635","language":[{"iso":"eng"}],"main_file_link":[{"open_access":"1","url":"http://arxiv.org/abs/2008.11995"}],"external_id":{"arxiv":["2008.11995"]},"oa":1,"quality_controlled":"1","article_number":"2180-2189","author":[{"full_name":"Royer, Amélie","orcid":"0000-0002-8407-0705","id":"3811D890-F248-11E8-B48F-1D18A9856A87","last_name":"Royer","first_name":"Amélie"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"related_material":{"record":[{"relation":"dissertation_contains","status":"deleted","id":"8331"},{"relation":"dissertation_contains","status":"public","id":"8390"}]},"date_updated":"2023-09-07T13:16:17Z","date_created":"2020-06-07T22:00:53Z","year":"2020","publication_status":"published","publisher":"IEEE","department":[{"_id":"ChLa"}],"day":"01","article_processing_charge":"No","scopus_import":"1","date_published":"2020-03-01T00:00:00Z","publication":"2020 IEEE Winter Conference on Applications of Computer Vision","citation":{"ama":"Royer A, Lampert C. A flexible selection scheme for minimum-effort transfer learning. In: 2020 IEEE Winter Conference on Applications of Computer Vision. IEEE; 2020. doi:10.1109/WACV45572.2020.9093635","ista":"Royer A, Lampert C. 2020. A flexible selection scheme for minimum-effort transfer learning. 2020 IEEE Winter Conference on Applications of Computer Vision. WACV: Winter Conference on Applications of Computer Vision, 2180–2189.","ieee":"A. Royer and C. Lampert, “A flexible selection scheme for minimum-effort transfer learning,” in 2020 IEEE Winter Conference on Applications of Computer Vision, Snowmass Village, CO, United States, 2020.","apa":"Royer, A., & Lampert, C. (2020). A flexible selection scheme for minimum-effort transfer learning. In 2020 IEEE Winter Conference on Applications of Computer Vision. Snowmass Village, CO, United States: IEEE. https://doi.org/10.1109/WACV45572.2020.9093635","mla":"Royer, Amélie, and Christoph Lampert. “A Flexible Selection Scheme for Minimum-Effort Transfer Learning.” 2020 IEEE Winter Conference on Applications of Computer Vision, 2180–2189, IEEE, 2020, doi:10.1109/WACV45572.2020.9093635.","short":"A. Royer, C. Lampert, in:, 2020 IEEE Winter Conference on Applications of Computer Vision, IEEE, 2020.","chicago":"Royer, Amélie, and Christoph Lampert. “A Flexible Selection Scheme for Minimum-Effort Transfer Learning.” In 2020 IEEE Winter Conference on Applications of Computer Vision. IEEE, 2020. https://doi.org/10.1109/WACV45572.2020.9093635."},"abstract":[{"text":"Fine-tuning is a popular way of exploiting knowledge contained in a pre-trained convolutional network for a new visual recognition task. However, the orthogonal setting of transferring knowledge from a pretrained network to a visually different yet semantically close source is rarely considered: This commonly happens with real-life data, which is not necessarily as clean as the training source (noise, geometric transformations, different modalities, etc.).To tackle such scenarios, we introduce a new, generalized form of fine-tuning, called flex-tuning, in which any individual unit (e.g. layer) of a network can be tuned, and the most promising one is chosen automatically. In order to make the method appealing for practical use, we propose two lightweight and faster selection procedures that prove to be good approximations in practice. We study these selection criteria empirically across a variety of domain shifts and data scarcity scenarios, and show that fine-tuning individual units, despite its simplicity, yields very good results as an adaptation technique. As it turns out, in contrast to common practice, rather than the last fully-connected unit it is best to tune an intermediate or early one in many domain- shift scenarios, which is accurately detected by flex-tuning.","lang":"eng"}],"type":"conference","oa_version":"Preprint","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"7937","title":"A flexible selection scheme for minimum-effort transfer learning","status":"public"},{"citation":{"chicago":"Royer, Amélie, Konstantinos Bousmalis, Stephan Gouws, Fred Bertsch, Inbar Mosseri, Forrester Cole, and Kevin Murphy. “XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings.” In Domain Adaptation for Visual Understanding, edited by Richa Singh, Mayank Vatsa, Vishal M. Patel, and Nalini Ratha, 33–49. Springer Nature, 2020. https://doi.org/10.1007/978-3-030-30671-7_3.","short":"A. Royer, K. Bousmalis, S. Gouws, F. Bertsch, I. Mosseri, F. Cole, K. Murphy, in:, R. Singh, M. Vatsa, V.M. Patel, N. Ratha (Eds.), Domain Adaptation for Visual Understanding, Springer Nature, 2020, pp. 33–49.","mla":"Royer, Amélie, et al. “XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings.” Domain Adaptation for Visual Understanding, edited by Richa Singh et al., Springer Nature, 2020, pp. 33–49, doi:10.1007/978-3-030-30671-7_3.","ieee":"A. Royer et al., “XGAN: Unsupervised image-to-image translation for many-to-many mappings,” in Domain Adaptation for Visual Understanding, R. Singh, M. Vatsa, V. M. Patel, and N. Ratha, Eds. Springer Nature, 2020, pp. 33–49.","apa":"Royer, A., Bousmalis, K., Gouws, S., Bertsch, F., Mosseri, I., Cole, F., & Murphy, K. (2020). XGAN: Unsupervised image-to-image translation for many-to-many mappings. In R. Singh, M. Vatsa, V. M. Patel, & N. Ratha (Eds.), Domain Adaptation for Visual Understanding (pp. 33–49). Springer Nature. https://doi.org/10.1007/978-3-030-30671-7_3","ista":"Royer A, Bousmalis K, Gouws S, Bertsch F, Mosseri I, Cole F, Murphy K. 2020.XGAN: Unsupervised image-to-image translation for many-to-many mappings. In: Domain Adaptation for Visual Understanding. , 33–49.","ama":"Royer A, Bousmalis K, Gouws S, et al. XGAN: Unsupervised image-to-image translation for many-to-many mappings. In: Singh R, Vatsa M, Patel VM, Ratha N, eds. Domain Adaptation for Visual Understanding. Springer Nature; 2020:33-49. doi:10.1007/978-3-030-30671-7_3"},"publication":"Domain Adaptation for Visual Understanding","page":"33-49","date_published":"2020-01-08T00:00:00Z","scopus_import":"1","article_processing_charge":"No","day":"08","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"8092","title":"XGAN: Unsupervised image-to-image translation for many-to-many mappings","status":"public","oa_version":"Preprint","type":"book_chapter","abstract":[{"text":"Image translation refers to the task of mapping images from a visual domain to another. Given two unpaired collections of images, we aim to learn a mapping between the corpus-level style of each collection, while preserving semantic content shared across the two domains. We introduce xgan, a dual adversarial auto-encoder, which captures a shared representation of the common domain semantic content in an unsupervised way, while jointly learning the domain-to-domain image translations in both directions. We exploit ideas from the domain adaptation literature and define a semantic consistency loss which encourages the learned embedding to preserve semantics shared across domains. We report promising qualitative results for the task of face-to-cartoon translation. The cartoon dataset we collected for this purpose, “CartoonSet”, is also publicly available as a new benchmark for semantic style transfer at https://google.github.io/cartoonset/index.html.","lang":"eng"}],"main_file_link":[{"url":"https://arxiv.org/abs/1711.05139","open_access":"1"}],"oa":1,"external_id":{"arxiv":["1711.05139"]},"quality_controlled":"1","doi":"10.1007/978-3-030-30671-7_3","language":[{"iso":"eng"}],"publication_identifier":{"isbn":["9783030306717"]},"month":"01","year":"2020","department":[{"_id":"ChLa"}],"editor":[{"first_name":"Richa","last_name":"Singh","full_name":"Singh, Richa"},{"full_name":"Vatsa, Mayank","first_name":"Mayank","last_name":"Vatsa"},{"full_name":"Patel, Vishal M.","first_name":"Vishal M.","last_name":"Patel"},{"full_name":"Ratha, Nalini","last_name":"Ratha","first_name":"Nalini"}],"publisher":"Springer Nature","publication_status":"published","related_material":{"record":[{"status":"deleted","relation":"dissertation_contains","id":"8331"},{"relation":"dissertation_contains","status":"public","id":"8390"}]},"author":[{"first_name":"Amélie","last_name":"Royer","id":"3811D890-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8407-0705","full_name":"Royer, Amélie"},{"first_name":"Konstantinos","last_name":"Bousmalis","full_name":"Bousmalis, Konstantinos"},{"full_name":"Gouws, Stephan","first_name":"Stephan","last_name":"Gouws"},{"last_name":"Bertsch","first_name":"Fred","full_name":"Bertsch, Fred"},{"full_name":"Mosseri, Inbar","last_name":"Mosseri","first_name":"Inbar"},{"first_name":"Forrester","last_name":"Cole","full_name":"Cole, Forrester"},{"first_name":"Kevin","last_name":"Murphy","full_name":"Murphy, Kevin"}],"date_updated":"2023-09-07T13:16:18Z","date_created":"2020-07-05T22:00:46Z"},{"article_processing_charge":"No","has_accepted_license":"1","day":"26","month":"04","language":[{"iso":"eng"}],"date_published":"2020-04-26T00:00:00Z","conference":{"start_date":"2020-04-27","location":"Online","end_date":"2020-04-30","name":"ICLR: International Conference on Learning Representations"},"quality_controlled":"1","citation":{"ieee":"M. Phuong and C. Lampert, “Functional vs. parametric equivalence of ReLU networks,” in 8th International Conference on Learning Representations, Online, 2020.","apa":"Phuong, M., & Lampert, C. (2020). Functional vs. parametric equivalence of ReLU networks. In 8th International Conference on Learning Representations. Online.","ista":"Phuong M, Lampert C. 2020. Functional vs. parametric equivalence of ReLU networks. 8th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","ama":"Phuong M, Lampert C. Functional vs. parametric equivalence of ReLU networks. In: 8th International Conference on Learning Representations. ; 2020.","chicago":"Phuong, Mary, and Christoph Lampert. “Functional vs. Parametric Equivalence of ReLU Networks.” In 8th International Conference on Learning Representations, 2020.","short":"M. Phuong, C. Lampert, in:, 8th International Conference on Learning Representations, 2020.","mla":"Phuong, Mary, and Christoph Lampert. “Functional vs. Parametric Equivalence of ReLU Networks.” 8th International Conference on Learning Representations, 2020."},"oa":1,"publication":"8th International Conference on Learning Representations","file_date_updated":"2020-07-14T12:47:59Z","abstract":[{"lang":"eng","text":"We address the following question: How redundant is the parameterisation of ReLU networks? Specifically, we consider transformations of the weight space which leave the function implemented by the network intact. Two such transformations are known for feed-forward architectures: permutation of neurons within a layer, and positive scaling of all incoming weights of a neuron coupled with inverse scaling of its outgoing weights. In this work, we show for architectures with non-increasing widths that permutation and scaling are in fact the only function-preserving weight transformations. For any eligible architecture we give an explicit construction of a neural network such that any other network that implements the same function can be obtained from the original one by the application of permutations and rescaling. The proof relies on a geometric understanding of boundaries between linear regions of ReLU networks, and we hope the developed mathematical tools are of independent interest."}],"type":"conference","oa_version":"Published Version","file":[{"access_level":"open_access","file_name":"main.pdf","creator":"bphuong","content_type":"application/pdf","file_size":405469,"file_id":"7482","relation":"main_file","checksum":"8d372ea5defd8cb8fdc430111ed754a9","date_updated":"2020-07-14T12:47:59Z","date_created":"2020-02-11T09:07:27Z"}],"date_created":"2020-02-11T09:07:37Z","date_updated":"2023-09-07T13:29:50Z","related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"9418"}],"link":[{"url":"https://iclr.cc/virtual_2020/poster_Bylx-TNKvH.html","relation":"supplementary_material"}]},"author":[{"full_name":"Bui Thi Mai, Phuong","first_name":"Phuong","last_name":"Bui Thi Mai","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"department":[{"_id":"ChLa"}],"status":"public","ddc":["000"],"title":"Functional vs. parametric equivalence of ReLU networks","publication_status":"published","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"7481","year":"2020"},{"file":[{"checksum":"cc755d0054bc4b2be778ea7aa7884d2f","success":1,"date_updated":"2021-02-15T09:00:01Z","date_created":"2021-02-15T09:00:01Z","relation":"main_file","file_id":"9120","file_size":281286,"content_type":"application/pdf","creator":"dernst","access_level":"open_access","file_name":"2020_PMLR_Konstantinov.pdf"}],"oa_version":"Published Version","intvolume":" 119","status":"public","title":"On the sample complexity of adversarial multi-source PAC learning","ddc":["000"],"user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","_id":"8724","abstract":[{"text":"We study the problem of learning from multiple untrusted data sources, a scenario of increasing practical relevance given the recent emergence of crowdsourcing and collaborative learning paradigms. Specifically, we analyze the situation in which a learning system obtains datasets from multiple sources, some of which might be biased or even adversarially perturbed. It is\r\nknown that in the single-source case, an adversary with the power to corrupt a fixed fraction of the training data can prevent PAC-learnability, that is, even in the limit of infinitely much training data, no learning system can approach the optimal test error. In this work we show that, surprisingly, the same is not true in the multi-source setting, where the adversary can arbitrarily\r\ncorrupt a fixed fraction of the data sources. Our main results are a generalization bound that provides finite-sample guarantees for this learning setting, as well as corresponding lower bounds. Besides establishing PAC-learnability our results also show that in a cooperative learning setting sharing data with other parties has provable benefits, even if some\r\nparticipants are malicious. ","lang":"eng"}],"type":"conference","date_published":"2020-07-12T00:00:00Z","page":"5416-5425","citation":{"short":"N.H. Konstantinov, E. Frantar, D.-A. Alistarh, C. Lampert, in:, Proceedings of the 37th International Conference on Machine Learning, ML Research Press, 2020, pp. 5416–5425.","mla":"Konstantinov, Nikola H., et al. “On the Sample Complexity of Adversarial Multi-Source PAC Learning.” Proceedings of the 37th International Conference on Machine Learning, vol. 119, ML Research Press, 2020, pp. 5416–25.","chicago":"Konstantinov, Nikola H, Elias Frantar, Dan-Adrian Alistarh, and Christoph Lampert. “On the Sample Complexity of Adversarial Multi-Source PAC Learning.” In Proceedings of the 37th International Conference on Machine Learning, 119:5416–25. ML Research Press, 2020.","ama":"Konstantinov NH, Frantar E, Alistarh D-A, Lampert C. On the sample complexity of adversarial multi-source PAC learning. In: Proceedings of the 37th International Conference on Machine Learning. Vol 119. ML Research Press; 2020:5416-5425.","ieee":"N. H. Konstantinov, E. Frantar, D.-A. Alistarh, and C. Lampert, “On the sample complexity of adversarial multi-source PAC learning,” in Proceedings of the 37th International Conference on Machine Learning, Online, 2020, vol. 119, pp. 5416–5425.","apa":"Konstantinov, N. H., Frantar, E., Alistarh, D.-A., & Lampert, C. (2020). On the sample complexity of adversarial multi-source PAC learning. In Proceedings of the 37th International Conference on Machine Learning (Vol. 119, pp. 5416–5425). Online: ML Research Press.","ista":"Konstantinov NH, Frantar E, Alistarh D-A, Lampert C. 2020. On the sample complexity of adversarial multi-source PAC learning. Proceedings of the 37th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 119, 5416–5425."},"publication":"Proceedings of the 37th International Conference on Machine Learning","article_processing_charge":"No","has_accepted_license":"1","day":"12","scopus_import":"1","volume":119,"date_created":"2020-11-05T15:25:58Z","date_updated":"2023-09-07T13:42:08Z","related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"10799"}],"link":[{"url":"http://proceedings.mlr.press/v119/konstantinov20a/konstantinov20a-supp.pdf","relation":"supplementary_material"}]},"author":[{"full_name":"Konstantinov, Nikola H","first_name":"Nikola H","last_name":"Konstantinov","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Elias","last_name":"Frantar","id":"09a8f98d-ec99-11ea-ae11-c063a7b7fe5f","full_name":"Frantar, Elias"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0003-3650-940X","first_name":"Dan-Adrian","last_name":"Alistarh","full_name":"Alistarh, Dan-Adrian"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"department":[{"_id":"DaAl"},{"_id":"ChLa"}],"publisher":"ML Research Press","publication_status":"published","acknowledgement":"Dan Alistarh is supported in part by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 805223 ScaleML). This research was supported by the Scientific Service Units (SSU) of IST Austria through resources provided by Scientific Computing (SciComp).","year":"2020","ec_funded":1,"file_date_updated":"2021-02-15T09:00:01Z","language":[{"iso":"eng"}],"acknowledged_ssus":[{"_id":"ScienComp"}],"conference":{"location":"Online","start_date":"2020-07-12","end_date":"2020-07-18","name":"ICML: International Conference on Machine Learning"},"project":[{"name":"Elastic Coordination for Scalable Machine Learning","call_identifier":"H2020","_id":"268A44D6-B435-11E9-9278-68D0E5697425","grant_number":"805223"}],"quality_controlled":"1","oa":1,"external_id":{"arxiv":["2002.10384"]},"publication_identifier":{"issn":["2640-3498"]},"month":"07"},{"language":[{"iso":"eng"}],"supervisor":[{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"}],"degree_awarded":"PhD","acknowledged_ssus":[{"_id":"CampIT"},{"_id":"ScienComp"}],"doi":"10.15479/AT:ISTA:8390","oa":1,"tmp":{"name":"Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)","legal_code_url":"https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode","image":"/images/cc_by_nc_sa.png","short":"CC BY-NC-SA (4.0)"},"publication_identifier":{"isbn":["978-3-99078-007-7"],"issn":["2663-337X"]},"month":"09","date_updated":"2023-10-16T10:04:02Z","date_created":"2020-09-14T13:42:09Z","related_material":{"record":[{"id":"7936","relation":"part_of_dissertation","status":"public"},{"id":"7937","relation":"part_of_dissertation","status":"public"},{"id":"8193","status":"public","relation":"part_of_dissertation"},{"status":"public","relation":"part_of_dissertation","id":"8092"},{"id":"911","status":"public","relation":"part_of_dissertation"}]},"author":[{"id":"3811D890-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8407-0705","first_name":"Amélie","last_name":"Royer","full_name":"Royer, Amélie"}],"publisher":"Institute of Science and Technology Austria","department":[{"_id":"ChLa"}],"publication_status":"published","acknowledgement":"Last but not least, I would like to acknowledge the support of the IST IT and scientific computing team for helping provide a great work environment.","year":"2020","license":"https://creativecommons.org/licenses/by-nc-sa/4.0/","file_date_updated":"2020-09-14T13:39:17Z","date_published":"2020-09-14T00:00:00Z","page":"197","citation":{"ama":"Royer A. Leveraging structure in Computer Vision tasks for flexible Deep Learning models. 2020. doi:10.15479/AT:ISTA:8390","ista":"Royer A. 2020. Leveraging structure in Computer Vision tasks for flexible Deep Learning models. Institute of Science and Technology Austria.","apa":"Royer, A. (2020). Leveraging structure in Computer Vision tasks for flexible Deep Learning models. Institute of Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:8390","ieee":"A. Royer, “Leveraging structure in Computer Vision tasks for flexible Deep Learning models,” Institute of Science and Technology Austria, 2020.","mla":"Royer, Amélie. Leveraging Structure in Computer Vision Tasks for Flexible Deep Learning Models. Institute of Science and Technology Austria, 2020, doi:10.15479/AT:ISTA:8390.","short":"A. Royer, Leveraging Structure in Computer Vision Tasks for Flexible Deep Learning Models, Institute of Science and Technology Austria, 2020.","chicago":"Royer, Amélie. “Leveraging Structure in Computer Vision Tasks for Flexible Deep Learning Models.” Institute of Science and Technology Austria, 2020. https://doi.org/10.15479/AT:ISTA:8390."},"article_processing_charge":"No","has_accepted_license":"1","day":"14","file":[{"content_type":"application/pdf","file_size":30224591,"creator":"dernst","access_level":"open_access","file_name":"2020_Thesis_Royer.pdf","checksum":"c914d2f88846032f3d8507734861b6ee","success":1,"date_updated":"2020-09-14T13:39:14Z","date_created":"2020-09-14T13:39:14Z","relation":"main_file","file_id":"8391"},{"access_level":"closed","file_name":"thesis_sources.zip","creator":"dernst","file_size":74227627,"content_type":"application/x-zip-compressed","file_id":"8392","relation":"main_file","checksum":"ae98fb35d912cff84a89035ae5794d3c","date_created":"2020-09-14T13:39:17Z","date_updated":"2020-09-14T13:39:17Z"}],"oa_version":"Published Version","ddc":["000"],"title":"Leveraging structure in Computer Vision tasks for flexible Deep Learning models","status":"public","_id":"8390","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","abstract":[{"text":"Deep neural networks have established a new standard for data-dependent feature extraction pipelines in the Computer Vision literature. Despite their remarkable performance in the standard supervised learning scenario, i.e. when models are trained with labeled data and tested on samples that follow a similar distribution, neural networks have been shown to struggle with more advanced generalization abilities, such as transferring knowledge across visually different domains, or generalizing to new unseen combinations of known concepts. In this thesis we argue that, in contrast to the usual black-box behavior of neural networks, leveraging more structured internal representations is a promising direction\r\nfor tackling such problems. In particular, we focus on two forms of structure. First, we tackle modularity: We show that (i) compositional architectures are a natural tool for modeling reasoning tasks, in that they efficiently capture their combinatorial nature, which is key for generalizing beyond the compositions seen during training. We investigate how to to learn such models, both formally and experimentally, for the task of abstract visual reasoning. Then, we show that (ii) in some settings, modularity allows us to efficiently break down complex tasks into smaller, easier, modules, thereby improving computational efficiency; We study this behavior in the context of generative models for colorization, as well as for small objects detection. Secondly, we investigate the inherently layered structure of representations learned by neural networks, and analyze its role in the context of transfer learning and domain adaptation across visually\r\ndissimilar domains. ","lang":"eng"}],"alternative_title":["ISTA Thesis"],"type":"dissertation"},{"file_date_updated":"2020-07-31T16:57:12Z","publisher":"IEEE","department":[{"_id":"ChLa"}],"publication_status":"published","year":"2020","date_created":"2020-07-31T16:53:49Z","date_updated":"2023-10-17T07:37:11Z","author":[{"full_name":"Henderson, Paul M","first_name":"Paul M","last_name":"Henderson","id":"13C09E74-18D9-11E9-8878-32CFE5697425","orcid":"0000-0002-5198-7445"},{"full_name":"Tsiminaki, Vagia","first_name":"Vagia","last_name":"Tsiminaki"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"}],"publication_identifier":{"eisbn":["9781728171685"],"eissn":["2575-7075"]},"month":"07","quality_controlled":"1","main_file_link":[{"open_access":"1","url":"https://openaccess.thecvf.com/content_CVPR_2020/papers/Henderson_Leveraging_2D_Data_to_Learn_Textured_3D_Mesh_Generation_CVPR_2020_paper.pdf"}],"oa":1,"external_id":{"arxiv":["2004.04180"]},"language":[{"iso":"eng"}],"doi":"10.1109/CVPR42600.2020.00752","conference":{"name":"CVPR: Conference on Computer Vision and Pattern Recognition","location":"Virtual","start_date":"2020-06-14","end_date":"2020-06-19"},"type":"conference","abstract":[{"text":"Numerous methods have been proposed for probabilistic generative modelling of\r\n3D objects. However, none of these is able to produce textured objects, which\r\nrenders them of limited use for practical tasks. In this work, we present the\r\nfirst generative model of textured 3D meshes. Training such a model would\r\ntraditionally require a large dataset of textured meshes, but unfortunately,\r\nexisting datasets of meshes lack detailed textures. We instead propose a new\r\ntraining methodology that allows learning from collections of 2D images without\r\nany 3D information. To do so, we train our model to explain a distribution of\r\nimages by modelling each image as a 3D foreground object placed in front of a\r\n2D background. Thus, it learns to generate meshes that when rendered, produce\r\nimages similar to those in its training set.\r\n A well-known problem when generating meshes with deep networks is the\r\nemergence of self-intersections, which are problematic for many use-cases. As a\r\nsecond contribution we therefore introduce a new generation process for 3D\r\nmeshes that guarantees no self-intersections arise, based on the physical\r\nintuition that faces should push one another out of the way as they move.\r\n We conduct extensive experiments on our approach, reporting quantitative and\r\nqualitative results on both synthetic data and natural images. These show our\r\nmethod successfully learns to generate plausible and diverse textured 3D\r\nsamples for five challenging object classes.","lang":"eng"}],"ddc":["004"],"title":"Leveraging 2D data to learn textured 3D mesh generation","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"8186","oa_version":"Submitted Version","file":[{"creator":"phenders","content_type":"application/pdf","file_size":10262773,"access_level":"open_access","file_name":"paper.pdf","success":1,"date_created":"2020-07-31T16:57:12Z","date_updated":"2020-07-31T16:57:12Z","file_id":"8187","relation":"main_file"}],"scopus_import":"1","article_processing_charge":"No","has_accepted_license":"1","day":"01","page":"7498-7507","citation":{"ista":"Henderson PM, Tsiminaki V, Lampert C. 2020. Leveraging 2D data to learn textured 3D mesh generation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Conference on Computer Vision and Pattern Recognition, 7498–7507.","apa":"Henderson, P. M., Tsiminaki, V., & Lampert, C. (2020). Leveraging 2D data to learn textured 3D mesh generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 7498–7507). Virtual: IEEE. https://doi.org/10.1109/CVPR42600.2020.00752","ieee":"P. M. Henderson, V. Tsiminaki, and C. Lampert, “Leveraging 2D data to learn textured 3D mesh generation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Virtual, 2020, pp. 7498–7507.","ama":"Henderson PM, Tsiminaki V, Lampert C. Leveraging 2D data to learn textured 3D mesh generation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE; 2020:7498-7507. doi:10.1109/CVPR42600.2020.00752","chicago":"Henderson, Paul M, Vagia Tsiminaki, and Christoph Lampert. “Leveraging 2D Data to Learn Textured 3D Mesh Generation.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7498–7507. IEEE, 2020. https://doi.org/10.1109/CVPR42600.2020.00752.","mla":"Henderson, Paul M., et al. “Leveraging 2D Data to Learn Textured 3D Mesh Generation.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2020, pp. 7498–507, doi:10.1109/CVPR42600.2020.00752.","short":"P.M. Henderson, V. Tsiminaki, C. Lampert, in:, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2020, pp. 7498–7507."},"publication":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition","date_published":"2020-07-01T00:00:00Z"},{"department":[{"_id":"ChLa"}],"publisher":"Springer Nature","publication_status":"published","year":"2020","volume":128,"date_updated":"2024-02-22T14:57:30Z","date_created":"2019-10-14T09:14:28Z","related_material":{"record":[{"relation":"earlier_version","status":"public","id":"6482"}],"link":[{"url":"https://doi.org/10.1007/s11263-019-01262-5","relation":"erratum"}]},"author":[{"full_name":"Sun, Rémy","last_name":"Sun","first_name":"Rémy"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"}],"ec_funded":1,"file_date_updated":"2020-07-14T12:47:45Z","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"},{"name":"IST Austria Open Access Fund","_id":"B67AFEDC-15C9-11EA-A837-991A96BB2854"}],"isi":1,"quality_controlled":"1","tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"oa":1,"external_id":{"isi":["000494406800001"]},"language":[{"iso":"eng"}],"doi":"10.1007/s11263-019-01232-x","publication_identifier":{"issn":["0920-5691"],"eissn":["1573-1405"]},"month":"04","intvolume":" 128","title":"KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications","status":"public","ddc":["004"],"_id":"6944","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","oa_version":"Published Version","file":[{"relation":"main_file","file_id":"7110","date_updated":"2020-07-14T12:47:45Z","date_created":"2019-11-26T10:30:02Z","checksum":"155e63edf664dcacb3bdc1c2223e606f","file_name":"2019_IJCV_Sun.pdf","access_level":"open_access","file_size":1715072,"content_type":"application/pdf","creator":"dernst"}],"type":"journal_article","issue":"4","abstract":[{"text":"We study the problem of automatically detecting if a given multi-class classifier operates outside of its specifications (out-of-specs), i.e. on input data from a different distribution than what it was trained for. This is an important problem to solve on the road towards creating reliable computer vision systems for real-world applications, because the quality of a classifier’s predictions cannot be guaranteed if it operates out-of-specs. Previously proposed methods for out-of-specs detection make decisions on the level of single inputs. This, however, is insufficient to achieve low false positive rate and high false negative rates at the same time. In this work, we describe a new procedure named KS(conf), based on statistical reasoning. Its main component is a classical Kolmogorov–Smirnov test that is applied to the set of predicted confidence values for batches of samples. Working with batches instead of single samples allows increasing the true positive rate without negatively affecting the false positive rate, thereby overcoming a crucial limitation of single sample tests. We show by extensive experiments using a variety of convolutional network architectures and datasets that KS(conf) reliably detects out-of-specs situations even under conditions where other tests fail. It furthermore has a number of properties that make it an excellent candidate for practical deployment: it is easy to implement, adds almost no overhead to the system, works with any classifier that outputs confidence scores, and requires no a priori knowledge about how the data distribution could change.","lang":"eng"}],"page":"970-995","article_type":"original","citation":{"chicago":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a Multiclass Classifier Operates Outside of Its Specifications.” International Journal of Computer Vision. Springer Nature, 2020. https://doi.org/10.1007/s11263-019-01232-x.","short":"R. Sun, C. Lampert, International Journal of Computer Vision 128 (2020) 970–995.","mla":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a Multiclass Classifier Operates Outside of Its Specifications.” International Journal of Computer Vision, vol. 128, no. 4, Springer Nature, 2020, pp. 970–95, doi:10.1007/s11263-019-01232-x.","apa":"Sun, R., & Lampert, C. (2020). KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. International Journal of Computer Vision. Springer Nature. https://doi.org/10.1007/s11263-019-01232-x","ieee":"R. Sun and C. Lampert, “KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications,” International Journal of Computer Vision, vol. 128, no. 4. Springer Nature, pp. 970–995, 2020.","ista":"Sun R, Lampert C. 2020. KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. International Journal of Computer Vision. 128(4), 970–995.","ama":"Sun R, Lampert C. KS(conf): A light-weight test if a multiclass classifier operates outside of its specifications. International Journal of Computer Vision. 2020;128(4):970-995. doi:10.1007/s11263-019-01232-x"},"publication":"International Journal of Computer Vision","date_published":"2020-04-01T00:00:00Z","scopus_import":"1","has_accepted_license":"1","article_processing_charge":"Yes (via OA deal)","day":"01"},{"citation":{"chicago":"Kersting, Kristian, Christoph Lampert, and Constantin Rothkopf, eds. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt. 1st ed. Wiesbaden: Springer Nature, 2019. https://doi.org/10.1007/978-3-658-26763-6.","short":"K. Kersting, C. Lampert, C. Rothkopf, eds., Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt, 1st ed., Springer Nature, Wiesbaden, 2019.","mla":"Kersting, Kristian, et al., editors. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt. 1st ed., Springer Nature, 2019, doi:10.1007/978-3-658-26763-6.","apa":"Kersting, K., Lampert, C., & Rothkopf, C. (Eds.). (2019). Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt (1st ed.). Wiesbaden: Springer Nature. https://doi.org/10.1007/978-3-658-26763-6","ieee":"K. Kersting, C. Lampert, and C. Rothkopf, Eds., Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt, 1st ed. Wiesbaden: Springer Nature, 2019.","ista":"Kersting K, Lampert C, Rothkopf C eds. 2019. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt 1st ed., Wiesbaden: Springer Nature, XIV, 245p.","ama":"Kersting K, Lampert C, Rothkopf C, eds. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt. 1st ed. Wiesbaden: Springer Nature; 2019. doi:10.1007/978-3-658-26763-6"},"page":"XIV, 245","quality_controlled":"1","date_published":"2019-10-30T00:00:00Z","doi":"10.1007/978-3-658-26763-6","language":[{"iso":"ger"}],"publication_identifier":{"eisbn":["978-3-658-26763-6"],"isbn":["978-3-658-26762-9"]},"article_processing_charge":"No","month":"10","day":"30","year":"2019","_id":"7171","user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","editor":[{"full_name":"Kersting, Kristian","last_name":"Kersting","first_name":"Kristian"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph"},{"last_name":"Rothkopf","first_name":"Constantin","full_name":"Rothkopf, Constantin"}],"department":[{"_id":"ChLa"}],"publisher":"Springer Nature","publication_status":"published","title":"Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt","status":"public","edition":"1","related_material":{"link":[{"relation":"press_release","description":"News on IST Website","url":"https://ist.ac.at/en/news/book-release-how-machines-learn/"}]},"oa_version":"None","date_updated":"2021-12-22T14:40:58Z","date_created":"2019-12-11T14:15:56Z","type":"book_editor","place":"Wiesbaden","abstract":[{"text":"Wissen Sie, was sich hinter künstlicher Intelligenz und maschinellem Lernen verbirgt? \r\nDieses Sachbuch erklärt Ihnen leicht verständlich und ohne komplizierte Formeln die grundlegenden Methoden und Vorgehensweisen des maschinellen Lernens. Mathematisches Vorwissen ist dafür nicht nötig. Kurzweilig und informativ illustriert Lisa, die Protagonistin des Buches, diese anhand von Alltagssituationen. \r\nEin Buch für alle, die in Diskussionen über Chancen und Risiken der aktuellen Entwicklung der künstlichen Intelligenz und des maschinellen Lernens mit Faktenwissen punkten möchten. Auch für Schülerinnen und Schüler geeignet!","lang":"ger"}]},{"status":"public","title":"Strategy representation by decision trees with linear classifiers","intvolume":" 11785","_id":"6942","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","oa_version":"Preprint","alternative_title":["LNCS"],"type":"conference","abstract":[{"lang":"eng","text":"Graph games and Markov decision processes (MDPs) are standard models in reactive synthesis and verification of probabilistic systems with nondeterminism. The class of 𝜔 -regular winning conditions; e.g., safety, reachability, liveness, parity conditions; provides a robust and expressive specification formalism for properties that arise in analysis of reactive systems. The resolutions of nondeterminism in games and MDPs are represented as strategies, and we consider succinct representation of such strategies. The decision-tree data structure from machine learning retains the flavor of decisions of strategies and allows entropy-based minimization to obtain succinct trees. However, in contrast to traditional machine-learning problems where small errors are allowed, for winning strategies in graph games and MDPs no error is allowed, and the decision tree must represent the entire strategy. In this work we propose decision trees with linear classifiers for representation of strategies in graph games and MDPs. We have implemented strategy representation using this data structure and we present experimental results for problems on graph games and MDPs, which show that this new data structure presents a much more efficient strategy representation as compared to standard decision trees."}],"page":"109-128","publication":"16th International Conference on Quantitative Evaluation of Systems","citation":{"ista":"Ashok P, Brázdil T, Chatterjee K, Křetínský J, Lampert C, Toman V. 2019. Strategy representation by decision trees with linear classifiers. 16th International Conference on Quantitative Evaluation of Systems. QEST: Quantitative Evaluation of Systems, LNCS, vol. 11785, 109–128.","ieee":"P. Ashok, T. Brázdil, K. Chatterjee, J. Křetínský, C. Lampert, and V. Toman, “Strategy representation by decision trees with linear classifiers,” in 16th International Conference on Quantitative Evaluation of Systems, Glasgow, United Kingdom, 2019, vol. 11785, pp. 109–128.","apa":"Ashok, P., Brázdil, T., Chatterjee, K., Křetínský, J., Lampert, C., & Toman, V. (2019). Strategy representation by decision trees with linear classifiers. In 16th International Conference on Quantitative Evaluation of Systems (Vol. 11785, pp. 109–128). Glasgow, United Kingdom: Springer Nature. https://doi.org/10.1007/978-3-030-30281-8_7","ama":"Ashok P, Brázdil T, Chatterjee K, Křetínský J, Lampert C, Toman V. Strategy representation by decision trees with linear classifiers. In: 16th International Conference on Quantitative Evaluation of Systems. Vol 11785. Springer Nature; 2019:109-128. doi:10.1007/978-3-030-30281-8_7","chicago":"Ashok, Pranav, Tomáš Brázdil, Krishnendu Chatterjee, Jan Křetínský, Christoph Lampert, and Viktor Toman. “Strategy Representation by Decision Trees with Linear Classifiers.” In 16th International Conference on Quantitative Evaluation of Systems, 11785:109–28. Springer Nature, 2019. https://doi.org/10.1007/978-3-030-30281-8_7.","mla":"Ashok, Pranav, et al. “Strategy Representation by Decision Trees with Linear Classifiers.” 16th International Conference on Quantitative Evaluation of Systems, vol. 11785, Springer Nature, 2019, pp. 109–28, doi:10.1007/978-3-030-30281-8_7.","short":"P. Ashok, T. Brázdil, K. Chatterjee, J. Křetínský, C. Lampert, V. Toman, in:, 16th International Conference on Quantitative Evaluation of Systems, Springer Nature, 2019, pp. 109–128."},"date_published":"2019-09-04T00:00:00Z","scopus_import":"1","day":"04","article_processing_charge":"No","publication_status":"published","department":[{"_id":"KrCh"},{"_id":"ChLa"}],"publisher":"Springer Nature","year":"2019","date_updated":"2023-08-30T06:59:36Z","date_created":"2019-10-14T06:57:49Z","volume":11785,"author":[{"full_name":"Ashok, Pranav","first_name":"Pranav","last_name":"Ashok"},{"last_name":"Brázdil","first_name":"Tomáš","full_name":"Brázdil, Tomáš"},{"first_name":"Krishnendu","last_name":"Chatterjee","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-4561-241X","full_name":"Chatterjee, Krishnendu"},{"full_name":"Křetínský, Jan","last_name":"Křetínský","first_name":"Jan"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"},{"orcid":"0000-0001-9036-063X","id":"3AF3DA7C-F248-11E8-B48F-1D18A9856A87","last_name":"Toman","first_name":"Viktor","full_name":"Toman, Viktor"}],"quality_controlled":"1","isi":1,"project":[{"_id":"25863FF4-B435-11E9-9278-68D0E5697425","grant_number":"S11407","name":"Game Theory","call_identifier":"FWF"},{"name":"Rigorous Systems Engineering","call_identifier":"FWF","_id":"25F2ACDE-B435-11E9-9278-68D0E5697425","grant_number":"S11402-N23"},{"name":"Efficient Algorithms for Computer Aided Verification","_id":"25892FC0-B435-11E9-9278-68D0E5697425","grant_number":"ICT15-003"}],"oa":1,"external_id":{"isi":["000679281300007"],"arxiv":["1906.08178"]},"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1906.08178"}],"language":[{"iso":"eng"}],"conference":{"name":"QEST: Quantitative Evaluation of Systems","end_date":"2019-09-12","start_date":"2019-09-10","location":"Glasgow, United Kingdom"},"doi":"10.1007/978-3-030-30281-8_7","month":"09","publication_identifier":{"isbn":["9783030302801"],"issn":["0302-9743"],"eisbn":["9783030302818"]}},{"scopus_import":"1","article_processing_charge":"No","day":"01","page":"2251 - 2265","article_type":"original","citation":{"ama":"Xian Y, Lampert C, Schiele B, Akata Z. Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2019;41(9):2251-2265. doi:10.1109/tpami.2018.2857768","ista":"Xian Y, Lampert C, Schiele B, Akata Z. 2019. Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly. IEEE Transactions on Pattern Analysis and Machine Intelligence. 41(9), 2251–2265.","apa":"Xian, Y., Lampert, C., Schiele, B., & Akata, Z. (2019). Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly. IEEE Transactions on Pattern Analysis and Machine Intelligence. Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/tpami.2018.2857768","ieee":"Y. Xian, C. Lampert, B. Schiele, and Z. Akata, “Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, no. 9. Institute of Electrical and Electronics Engineers (IEEE), pp. 2251–2265, 2019.","mla":"Xian, Yongqin, et al. “Zero-Shot Learning - A Comprehensive Evaluation of the Good, the Bad and the Ugly.” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, no. 9, Institute of Electrical and Electronics Engineers (IEEE), 2019, pp. 2251–65, doi:10.1109/tpami.2018.2857768.","short":"Y. Xian, C. Lampert, B. Schiele, Z. Akata, IEEE Transactions on Pattern Analysis and Machine Intelligence 41 (2019) 2251–2265.","chicago":"Xian, Yongqin, Christoph Lampert, Bernt Schiele, and Zeynep Akata. “Zero-Shot Learning - A Comprehensive Evaluation of the Good, the Bad and the Ugly.” IEEE Transactions on Pattern Analysis and Machine Intelligence. Institute of Electrical and Electronics Engineers (IEEE), 2019. https://doi.org/10.1109/tpami.2018.2857768."},"publication":"IEEE Transactions on Pattern Analysis and Machine Intelligence","date_published":"2019-09-01T00:00:00Z","type":"journal_article","issue":"9","abstract":[{"lang":"eng","text":"Due to the importance of zero-shot learning, i.e. classifying images where there is a lack of labeled training data, the number of proposed approaches has recently increased steadily. We argue that it is time to take a step back and to analyze the status quo of the area. The purpose of this paper is three-fold. First, given the fact that there is no agreed upon zero-shot learning benchmark, we first define a new benchmark by unifying both the evaluation protocols and data splits of publicly available datasets used for this task. This is an important contribution as published results are often not comparable and sometimes even flawed due to, e.g. pre-training on zero-shot test classes. Moreover, we propose a new zero-shot learning dataset, the Animals with Attributes 2 (AWA2) dataset which we make publicly available both in terms of image features and the images themselves. Second, we compare and analyze a significant number of the state-of-the-art methods in depth, both in the classic zero-shot setting but also in the more realistic generalized zero-shot setting. Finally, we discuss in detail the limitations of the current status of the area which can be taken as a basis for advancing it."}],"intvolume":" 41","status":"public","title":"Zero-shot learning - A comprehensive evaluation of the good, the bad and the ugly","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"6554","oa_version":"Preprint","publication_identifier":{"issn":["0162-8828"],"eissn":["1939-3539"]},"month":"09","quality_controlled":"1","isi":1,"main_file_link":[{"url":"https://arxiv.org/abs/1707.00600","open_access":"1"}],"external_id":{"isi":["000480343900015"],"arxiv":["1707.00600"]},"oa":1,"language":[{"iso":"eng"}],"doi":"10.1109/tpami.2018.2857768","department":[{"_id":"ChLa"}],"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","publication_status":"published","year":"2019","volume":41,"date_created":"2019-06-11T14:05:59Z","date_updated":"2023-09-05T13:18:09Z","author":[{"first_name":"Yongqin","last_name":"Xian","full_name":"Xian, Yongqin"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-4561-241X","full_name":"Lampert, Christoph"},{"full_name":"Schiele, Bernt","last_name":"Schiele","first_name":"Bernt"},{"full_name":"Akata, Zeynep","last_name":"Akata","first_name":"Zeynep"}]},{"file_date_updated":"2020-07-14T12:47:59Z","ec_funded":1,"author":[{"full_name":"Bui Thi Mai, Phuong","last_name":"Bui Thi Mai","first_name":"Phuong","id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"related_material":{"record":[{"id":"9418","relation":"dissertation_contains","status":"public"}]},"date_updated":"2023-09-08T11:11:12Z","date_created":"2020-02-11T09:06:57Z","volume":"2019-October","year":"2019","publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"IEEE","month":"10","publication_identifier":{"isbn":["9781728148038"],"issn":["15505499"]},"conference":{"name":"ICCV: International Conference on Computer Vision","end_date":"2019-11-02","start_date":"2019-10-27","location":"Seoul, Korea"},"doi":"10.1109/ICCV.2019.00144","language":[{"iso":"eng"}],"external_id":{"isi":["000531438101047"]},"oa":1,"isi":1,"quality_controlled":"1","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"abstract":[{"lang":"eng","text":"Multi-exit architectures, in which a stack of processing layers is interleaved with early output layers, allow the processing of a test example to stop early and thus save computation time and/or energy. In this work, we propose a new training procedure for multi-exit architectures based on the principle of knowledge distillation. The method encourage searly exits to mimic later, more accurate exits, by matching their output probabilities.\r\nExperiments on CIFAR100 and ImageNet show that distillation-based training significantly improves the accuracy of early exits while maintaining state-of-the-art accuracy for late ones. The method is particularly beneficial when training data is limited and it allows a straightforward extension to semi-supervised learning,i.e. making use of unlabeled data at training time. Moreover, it takes only afew lines to implement and incurs almost no computational overhead at training time, and none at all at test time."}],"type":"conference","oa_version":"Submitted Version","file":[{"file_name":"main.pdf","access_level":"open_access","creator":"bphuong","file_size":735768,"content_type":"application/pdf","file_id":"7480","relation":"main_file","date_created":"2020-02-11T09:06:39Z","date_updated":"2020-07-14T12:47:59Z","checksum":"7b77fb5c2d27c4c37a7612ba46a66117"}],"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"7479","status":"public","title":"Distillation-based training for multi-exit architectures","ddc":["000"],"day":"01","has_accepted_license":"1","article_processing_charge":"No","scopus_import":"1","date_published":"2019-10-01T00:00:00Z","publication":"IEEE International Conference on Computer Vision","citation":{"short":"M. Phuong, C. Lampert, in:, IEEE International Conference on Computer Vision, IEEE, 2019, pp. 1355–1364.","mla":"Phuong, Mary, and Christoph Lampert. “Distillation-Based Training for Multi-Exit Architectures.” IEEE International Conference on Computer Vision, vol. 2019–October, IEEE, 2019, pp. 1355–64, doi:10.1109/ICCV.2019.00144.","chicago":"Phuong, Mary, and Christoph Lampert. “Distillation-Based Training for Multi-Exit Architectures.” In IEEE International Conference on Computer Vision, 2019–October:1355–64. IEEE, 2019. https://doi.org/10.1109/ICCV.2019.00144.","ama":"Phuong M, Lampert C. Distillation-based training for multi-exit architectures. In: IEEE International Conference on Computer Vision. Vol 2019-October. IEEE; 2019:1355-1364. doi:10.1109/ICCV.2019.00144","apa":"Phuong, M., & Lampert, C. (2019). Distillation-based training for multi-exit architectures. In IEEE International Conference on Computer Vision (Vol. 2019–October, pp. 1355–1364). Seoul, Korea: IEEE. https://doi.org/10.1109/ICCV.2019.00144","ieee":"M. Phuong and C. Lampert, “Distillation-based training for multi-exit architectures,” in IEEE International Conference on Computer Vision, Seoul, Korea, 2019, vol. 2019–October, pp. 1355–1364.","ista":"Phuong M, Lampert C. 2019. Distillation-based training for multi-exit architectures. IEEE International Conference on Computer Vision. ICCV: International Conference on Computer Vision vol. 2019–October, 1355–1364."},"page":"1355-1364"},{"article_number":"1749-1753","ec_funded":1,"department":[{"_id":"ChLa"}],"publisher":"IEEE","publication_status":"published","year":"2019","date_updated":"2023-09-08T11:18:37Z","date_created":"2020-04-05T22:00:51Z","author":[{"id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","first_name":"Alexander","last_name":"Kolesnikov","full_name":"Kolesnikov, Alexander"},{"last_name":"Kuznetsova","first_name":"Alina","full_name":"Kuznetsova, Alina"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"},{"first_name":"Vittorio","last_name":"Ferrari","full_name":"Ferrari, Vittorio"}],"publication_identifier":{"isbn":["9781728150239"]},"month":"10","project":[{"grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425","call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding"}],"quality_controlled":"1","isi":1,"oa":1,"external_id":{"arxiv":["1807.02136"],"isi":["000554591601098"]},"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1807.02136"}],"language":[{"iso":"eng"}],"doi":"10.1109/ICCVW.2019.00217","conference":{"end_date":"2019-10-28","start_date":"2019-10-27","location":"Seoul, South Korea","name":"ICCVW: International Conference on Computer Vision Workshop"},"type":"conference","abstract":[{"lang":"eng","text":"We propose a new model for detecting visual relationships, such as \"person riding motorcycle\" or \"bottle on table\". This task is an important step towards comprehensive structured mage understanding, going beyond detecting individual objects. Our main novelty is a Box Attention mechanism that allows to model pairwise interactions between objects using standard object detection pipelines. The resulting model is conceptually clean, expressive and relies on well-justified training and prediction procedures. Moreover, unlike previously proposed approaches, our model does not introduce any additional complex components or hyperparameters on top of those already required by the underlying detection model. We conduct an experimental evaluation on two datasets, V-COCO and Open Images, demonstrating strong quantitative and qualitative results."}],"title":"Detecting visual relationships using box attention","status":"public","_id":"7640","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","oa_version":"Preprint","scopus_import":"1","article_processing_charge":"No","day":"01","citation":{"mla":"Kolesnikov, Alexander, et al. “Detecting Visual Relationships Using Box Attention.” Proceedings of the 2019 International Conference on Computer Vision Workshop, 1749–1753, IEEE, 2019, doi:10.1109/ICCVW.2019.00217.","short":"A. Kolesnikov, A. Kuznetsova, C. Lampert, V. Ferrari, in:, Proceedings of the 2019 International Conference on Computer Vision Workshop, IEEE, 2019.","chicago":"Kolesnikov, Alexander, Alina Kuznetsova, Christoph Lampert, and Vittorio Ferrari. “Detecting Visual Relationships Using Box Attention.” In Proceedings of the 2019 International Conference on Computer Vision Workshop. IEEE, 2019. https://doi.org/10.1109/ICCVW.2019.00217.","ama":"Kolesnikov A, Kuznetsova A, Lampert C, Ferrari V. Detecting visual relationships using box attention. In: Proceedings of the 2019 International Conference on Computer Vision Workshop. IEEE; 2019. doi:10.1109/ICCVW.2019.00217","ista":"Kolesnikov A, Kuznetsova A, Lampert C, Ferrari V. 2019. Detecting visual relationships using box attention. Proceedings of the 2019 International Conference on Computer Vision Workshop. ICCVW: International Conference on Computer Vision Workshop, 1749–1753.","ieee":"A. Kolesnikov, A. Kuznetsova, C. Lampert, and V. Ferrari, “Detecting visual relationships using box attention,” in Proceedings of the 2019 International Conference on Computer Vision Workshop, Seoul, South Korea, 2019.","apa":"Kolesnikov, A., Kuznetsova, A., Lampert, C., & Ferrari, V. (2019). Detecting visual relationships using box attention. In Proceedings of the 2019 International Conference on Computer Vision Workshop. Seoul, South Korea: IEEE. https://doi.org/10.1109/ICCVW.2019.00217"},"publication":"Proceedings of the 2019 International Conference on Computer Vision Workshop","date_published":"2019-10-01T00:00:00Z"},{"type":"conference","abstract":[{"lang":"eng","text":"Knowledge distillation, i.e. one classifier being trained on the outputs of another classifier, is an empirically very successful technique for knowledge transfer between classifiers. It has even been observed that classifiers learn much faster and more reliably if trained with the outputs of another classifier as soft labels, instead of from ground truth data. So far, however, there is no satisfactory theoretical explanation of this phenomenon. In this work, we provide the first insights into the working mechanisms of distillation by studying the special case of linear and deep linear classifiers. Specifically, we prove a generalization bound that establishes fast convergence of the expected risk of a distillation-trained linear classifier. From the bound and its proof we extract three keyfactors that determine the success of distillation: data geometry – geometric properties of the datadistribution, in particular class separation, has an immediate influence on the convergence speed of the risk; optimization bias– gradient descentoptimization finds a very favorable minimum of the distillation objective; and strong monotonicity– the expected risk of the student classifier always decreases when the size of the training set grows."}],"_id":"6569","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","title":"Towards understanding knowledge distillation","ddc":["000"],"status":"public","intvolume":" 97","file":[{"date_updated":"2020-07-14T12:47:33Z","date_created":"2019-06-20T18:22:56Z","checksum":"a66d00e2694d749250f8507f301320ca","relation":"main_file","file_id":"6570","file_size":686432,"content_type":"application/pdf","creator":"bphuong","file_name":"paper.pdf","access_level":"open_access"}],"oa_version":"Published Version","scopus_import":"1","day":"13","has_accepted_license":"1","article_processing_charge":"No","publication":"Proceedings of the 36th International Conference on Machine Learning","citation":{"short":"M. Phuong, C. Lampert, in:, Proceedings of the 36th International Conference on Machine Learning, ML Research Press, 2019, pp. 5142–5151.","mla":"Phuong, Mary, and Christoph Lampert. “Towards Understanding Knowledge Distillation.” Proceedings of the 36th International Conference on Machine Learning, vol. 97, ML Research Press, 2019, pp. 5142–51.","chicago":"Phuong, Mary, and Christoph Lampert. “Towards Understanding Knowledge Distillation.” In Proceedings of the 36th International Conference on Machine Learning, 97:5142–51. ML Research Press, 2019.","ama":"Phuong M, Lampert C. Towards understanding knowledge distillation. In: Proceedings of the 36th International Conference on Machine Learning. Vol 97. ML Research Press; 2019:5142-5151.","apa":"Phuong, M., & Lampert, C. (2019). Towards understanding knowledge distillation. In Proceedings of the 36th International Conference on Machine Learning (Vol. 97, pp. 5142–5151). Long Beach, CA, United States: ML Research Press.","ieee":"M. Phuong and C. Lampert, “Towards understanding knowledge distillation,” in Proceedings of the 36th International Conference on Machine Learning, Long Beach, CA, United States, 2019, vol. 97, pp. 5142–5151.","ista":"Phuong M, Lampert C. 2019. Towards understanding knowledge distillation. Proceedings of the 36th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 97, 5142–5151."},"page":"5142-5151","date_published":"2019-06-13T00:00:00Z","file_date_updated":"2020-07-14T12:47:33Z","year":"2019","publication_status":"published","publisher":"ML Research Press","department":[{"_id":"ChLa"}],"author":[{"id":"3EC6EE64-F248-11E8-B48F-1D18A9856A87","first_name":"Phuong","last_name":"Bui Thi Mai","full_name":"Bui Thi Mai, Phuong"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2019-06-20T18:23:03Z","date_updated":"2023-10-17T12:31:38Z","volume":97,"month":"06","oa":1,"quality_controlled":"1","conference":{"name":"ICML: International Conference on Machine Learning","end_date":"2019-06-15","location":"Long Beach, CA, United States","start_date":"2019-06-10"},"language":[{"iso":"eng"}]},{"date_published":"2019-06-01T00:00:00Z","page":"3488-3498","publication":"Proceedings of the 36th International Conference on Machine Learning","citation":{"ama":"Konstantinov NH, Lampert C. Robust learning from untrusted sources. In: Proceedings of the 36th International Conference on Machine Learning. Vol 97. ML Research Press; 2019:3488-3498.","ista":"Konstantinov NH, Lampert C. 2019. Robust learning from untrusted sources. Proceedings of the 36th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 97, 3488–3498.","apa":"Konstantinov, N. H., & Lampert, C. (2019). Robust learning from untrusted sources. In Proceedings of the 36th International Conference on Machine Learning (Vol. 97, pp. 3488–3498). Long Beach, CA, USA: ML Research Press.","ieee":"N. H. Konstantinov and C. Lampert, “Robust learning from untrusted sources,” in Proceedings of the 36th International Conference on Machine Learning, Long Beach, CA, USA, 2019, vol. 97, pp. 3488–3498.","mla":"Konstantinov, Nikola H., and Christoph Lampert. “Robust Learning from Untrusted Sources.” Proceedings of the 36th International Conference on Machine Learning, vol. 97, ML Research Press, 2019, pp. 3488–98.","short":"N.H. Konstantinov, C. Lampert, in:, Proceedings of the 36th International Conference on Machine Learning, ML Research Press, 2019, pp. 3488–3498.","chicago":"Konstantinov, Nikola H, and Christoph Lampert. “Robust Learning from Untrusted Sources.” In Proceedings of the 36th International Conference on Machine Learning, 97:3488–98. ML Research Press, 2019."},"day":"01","article_processing_charge":"No","scopus_import":"1","oa_version":"Preprint","status":"public","title":"Robust learning from untrusted sources","intvolume":" 97","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"6590","abstract":[{"text":"Modern machine learning methods often require more data for training than a single expert can provide. Therefore, it has become a standard procedure to collect data from external sources, e.g. via crowdsourcing. Unfortunately, the quality of these sources is not always guaranteed. As additional complications, the data might be stored in a distributed way, or might even have to remain private. In this work, we address the question of how to learn robustly in such scenarios. Studying the problem through the lens of statistical learning theory, we derive a procedure that allows for learning from all available sources, yet automatically suppresses irrelevant or corrupted data. We show by extensive experiments that our method provides significant improvements over alternative approaches from robust statistics and distributed optimization. ","lang":"eng"}],"type":"conference","language":[{"iso":"eng"}],"conference":{"start_date":"2019-06-10","location":"Long Beach, CA, USA","end_date":"2919-06-15","name":"ICML: International Conference on Machine Learning"},"quality_controlled":"1","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"},{"name":"International IST Doctoral Program","call_identifier":"H2020","grant_number":"665385","_id":"2564DBCA-B435-11E9-9278-68D0E5697425"}],"oa":1,"main_file_link":[{"url":"https://arxiv.org/abs/1901.10310","open_access":"1"}],"external_id":{"arxiv":["1901.10310"]},"month":"06","date_created":"2019-06-27T14:18:23Z","date_updated":"2023-10-17T12:31:55Z","volume":97,"author":[{"id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","last_name":"Konstantinov","first_name":"Nikola H","full_name":"Konstantinov, Nikola H"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"related_material":{"record":[{"id":"10799","relation":"dissertation_contains","status":"public"}]},"publication_status":"published","publisher":"ML Research Press","department":[{"_id":"ChLa"}],"year":"2019","ec_funded":1},{"abstract":[{"text":"Computer vision systems for automatic image categorization have become accurate and reliable enough that they can run continuously for days or even years as components of real-world commercial applications. A major open problem in this context, however, is quality control. Good classification performance can only be expected if systems run under the specific conditions, in particular data distributions, that they were trained for. Surprisingly, none of the currently used deep network architectures have a built-in functionality that could detect if a network operates on data from a distribution it was not trained for, such that potentially a warning to the human users could be triggered. In this work, we describe KS(conf), a procedure for detecting such outside of specifications (out-of-specs) operation, based on statistical testing of the network outputs. We show by extensive experiments using the ImageNet, AwA2 and DAVIS datasets on a variety of ConvNets architectures that KS(conf) reliably detects out-of-specs situations. It furthermore has a number of properties that make it a promising candidate for practical deployment: it is easy to implement, adds almost no overhead to the system, works with all networks, including pretrained ones, and requires no a priori knowledge of how the data distribution could change. ","lang":"eng"}],"alternative_title":["LNCS"],"type":"conference","oa_version":"Preprint","status":"public","title":"KS(conf): A light-weight test if a ConvNet operates outside of Its specifications","intvolume":" 11269","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"6482","day":"14","article_processing_charge":"No","scopus_import":"1","date_published":"2019-02-14T00:00:00Z","page":"244-259","citation":{"ista":"Sun R, Lampert C. 2019. KS(conf): A light-weight test if a ConvNet operates outside of Its specifications. GCPR: Conference on Pattern Recognition, LNCS, vol. 11269, 244–259.","apa":"Sun, R., & Lampert, C. (2019). KS(conf): A light-weight test if a ConvNet operates outside of Its specifications (Vol. 11269, pp. 244–259). Presented at the GCPR: Conference on Pattern Recognition, Stuttgart, Germany: Springer Nature. https://doi.org/10.1007/978-3-030-12939-2_18","ieee":"R. Sun and C. Lampert, “KS(conf): A light-weight test if a ConvNet operates outside of Its specifications,” presented at the GCPR: Conference on Pattern Recognition, Stuttgart, Germany, 2019, vol. 11269, pp. 244–259.","ama":"Sun R, Lampert C. KS(conf): A light-weight test if a ConvNet operates outside of Its specifications. In: Vol 11269. Springer Nature; 2019:244-259. doi:10.1007/978-3-030-12939-2_18","chicago":"Sun, Rémy, and Christoph Lampert. “KS(Conf): A Light-Weight Test If a ConvNet Operates Outside of Its Specifications,” 11269:244–59. Springer Nature, 2019. https://doi.org/10.1007/978-3-030-12939-2_18.","mla":"Sun, Rémy, and Christoph Lampert. KS(Conf): A Light-Weight Test If a ConvNet Operates Outside of Its Specifications. Vol. 11269, Springer Nature, 2019, pp. 244–59, doi:10.1007/978-3-030-12939-2_18.","short":"R. Sun, C. Lampert, in:, Springer Nature, 2019, pp. 244–259."},"ec_funded":1,"date_created":"2019-05-24T09:48:36Z","date_updated":"2024-02-22T14:57:29Z","volume":11269,"author":[{"last_name":"Sun","first_name":"Rémy","full_name":"Sun, Rémy"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"related_material":{"record":[{"status":"public","relation":"later_version","id":"6944"}]},"publication_status":"published","publisher":"Springer Nature","department":[{"_id":"ChLa"}],"year":"2019","month":"02","publication_identifier":{"isbn":["9783030129385","9783030129392"],"eissn":["1611-3349"],"issn":["0302-9743"]},"language":[{"iso":"eng"}],"conference":{"name":"GCPR: Conference on Pattern Recognition","end_date":"2018-10-12","location":"Stuttgart, Germany","start_date":"2018-10-09"},"doi":"10.1007/978-3-030-12939-2_18","quality_controlled":"1","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"oa":1,"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1804.04171"}],"external_id":{"arxiv":["1804.04171"]}},{"author":[{"full_name":"Zimin, Alexander","first_name":"Alexander","last_name":"Zimin","id":"37099E9C-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2018-12-11T11:44:27Z","date_updated":"2023-09-07T12:29:07Z","year":"2018","publisher":"Institute of Science and Technology Austria","department":[{"_id":"ChLa"}],"publication_status":"published","publist_id":"7986","ec_funded":1,"file_date_updated":"2020-07-14T12:47:40Z","doi":"10.15479/AT:ISTA:TH1048","language":[{"iso":"eng"}],"degree_awarded":"PhD","supervisor":[{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"oa":1,"project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"publication_identifier":{"issn":["2663-337X"]},"month":"09","pubrep_id":"1048","file":[{"file_id":"6253","relation":"main_file","checksum":"e849dd40a915e4d6c5572b51b517f098","date_updated":"2020-07-14T12:47:40Z","date_created":"2019-04-09T07:32:47Z","access_level":"open_access","file_name":"2018_Thesis_Zimin.pdf","creator":"dernst","content_type":"application/pdf","file_size":1036137},{"file_name":"2018_Thesis_Zimin_Source.zip","access_level":"closed","creator":"dernst","content_type":"application/zip","file_size":637490,"file_id":"6254","relation":"source_file","date_updated":"2020-07-14T12:47:40Z","date_created":"2019-04-09T07:32:47Z","checksum":"da092153cec55c97461bd53c45c5d139"}],"oa_version":"Published Version","_id":"68","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","title":"Learning from dependent data","status":"public","ddc":["004","519"],"abstract":[{"lang":"eng","text":"The most common assumption made in statistical learning theory is the assumption of the independent and identically distributed (i.i.d.) data. While being very convenient mathematically, it is often very clearly violated in practice. This disparity between the machine learning theory and applications underlies a growing demand in the development of algorithms that learn from dependent data and theory that can provide generalization guarantees similar to the independent situations. This thesis is dedicated to two variants of dependencies that can arise in practice. One is a dependence on the level of samples in a single learning task. Another dependency type arises in the multi-task setting when the tasks are dependent on each other even though the data for them can be i.i.d. In both cases we model the data (samples or tasks) as stochastic processes and introduce new algorithms for both settings that take into account and exploit the resulting dependencies. We prove the theoretical guarantees on the performance of the introduced algorithms under different evaluation criteria and, in addition, we compliment the theoretical study by the empirical one, where we evaluate some of the algorithms on two real world datasets to highlight their practical applicability."}],"type":"dissertation","alternative_title":["ISTA Thesis"],"date_published":"2018-09-01T00:00:00Z","citation":{"ista":"Zimin A. 2018. Learning from dependent data. Institute of Science and Technology Austria.","apa":"Zimin, A. (2018). Learning from dependent data. Institute of Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:TH1048","ieee":"A. Zimin, “Learning from dependent data,” Institute of Science and Technology Austria, 2018.","ama":"Zimin A. Learning from dependent data. 2018. doi:10.15479/AT:ISTA:TH1048","chicago":"Zimin, Alexander. “Learning from Dependent Data.” Institute of Science and Technology Austria, 2018. https://doi.org/10.15479/AT:ISTA:TH1048.","mla":"Zimin, Alexander. Learning from Dependent Data. Institute of Science and Technology Austria, 2018, doi:10.15479/AT:ISTA:TH1048.","short":"A. Zimin, Learning from Dependent Data, Institute of Science and Technology Austria, 2018."},"page":"92","has_accepted_license":"1","article_processing_charge":"No","day":"01"},{"citation":{"short":"A. Kolesnikov, Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images, Institute of Science and Technology Austria, 2018.","mla":"Kolesnikov, Alexander. Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images. Institute of Science and Technology Austria, 2018, doi:10.15479/AT:ISTA:th_1021.","chicago":"Kolesnikov, Alexander. “Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images.” Institute of Science and Technology Austria, 2018. https://doi.org/10.15479/AT:ISTA:th_1021.","ama":"Kolesnikov A. Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images. 2018. doi:10.15479/AT:ISTA:th_1021","apa":"Kolesnikov, A. (2018). Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images. Institute of Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:th_1021","ieee":"A. Kolesnikov, “Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images,” Institute of Science and Technology Austria, 2018.","ista":"Kolesnikov A. 2018. Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images. Institute of Science and Technology Austria."},"page":"113","date_published":"2018-05-25T00:00:00Z","article_processing_charge":"No","has_accepted_license":"1","day":"25","_id":"197","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","ddc":["004"],"status":"public","title":"Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images","pubrep_id":"1021","file":[{"content_type":"application/pdf","file_size":12918758,"creator":"system","access_level":"open_access","file_name":"IST-2018-1021-v1+1_thesis-unsigned-pdfa.pdf","checksum":"bc678e02468d8ebc39dc7267dfb0a1c4","date_created":"2018-12-12T10:14:57Z","date_updated":"2020-07-14T12:45:22Z","relation":"main_file","file_id":"5113"},{"file_name":"2018_Thesis_Kolesnikov_source.zip","access_level":"closed","creator":"dernst","content_type":"application/zip","file_size":55973760,"file_id":"6225","relation":"source_file","date_updated":"2020-07-14T12:45:22Z","date_created":"2019-04-05T09:34:49Z","checksum":"bc66973b086da5a043f1162dcfb1fde4"}],"oa_version":"Published Version","type":"dissertation","alternative_title":["ISTA Thesis"],"abstract":[{"lang":"eng","text":"Modern computer vision systems heavily rely on statistical machine learning models, which typically require large amounts of labeled data to be learned reliably. Moreover, very recently computer vision research widely adopted techniques for representation learning, which further increase the demand for labeled data. However, for many important practical problems there is relatively small amount of labeled data available, so it is problematic to leverage full potential of the representation learning methods. One way to overcome this obstacle is to invest substantial resources into producing large labelled datasets. Unfortunately, this can be prohibitively expensive in practice. In this thesis we focus on the alternative way of tackling the aforementioned issue. We concentrate on methods, which make use of weakly-labeled or even unlabeled data. Specifically, the first half of the thesis is dedicated to the semantic image segmentation task. We develop a technique, which achieves competitive segmentation performance and only requires annotations in a form of global image-level labels instead of dense segmentation masks. Subsequently, we present a new methodology, which further improves segmentation performance by leveraging tiny additional feedback from a human annotator. By using our methods practitioners can greatly reduce the amount of data annotation effort, which is required to learn modern image segmentation models. In the second half of the thesis we focus on methods for learning from unlabeled visual data. We study a family of autoregressive models for modeling structure of natural images and discuss potential applications of these models. Moreover, we conduct in-depth study of one of these applications, where we develop the state-of-the-art model for the probabilistic image colorization task."}],"oa":1,"project":[{"_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036","name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7"}],"doi":"10.15479/AT:ISTA:th_1021","language":[{"iso":"eng"}],"degree_awarded":"PhD","supervisor":[{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"publication_identifier":{"issn":["2663-337X"]},"month":"05","year":"2018","acknowledgement":"I also gratefully acknowledge the support of NVIDIA Corporation with the donation of the GPUs used for this research.","department":[{"_id":"ChLa"}],"publisher":"Institute of Science and Technology Austria","publication_status":"published","author":[{"id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","last_name":"Kolesnikov","first_name":"Alexander","full_name":"Kolesnikov, Alexander"}],"date_created":"2018-12-11T11:45:09Z","date_updated":"2023-09-07T12:51:46Z","publist_id":"7718","ec_funded":1,"file_date_updated":"2020-07-14T12:45:22Z"},{"citation":{"short":"H. Ringbauer, A. Kolesnikov, D. Field, N.H. Barton, Genetics 208 (2018) 1231–1245.","mla":"Ringbauer, Harald, et al. “Estimating Barriers to Gene Flow from Distorted Isolation-by-Distance Patterns.” Genetics, vol. 208, no. 3, Genetics Society of America, 2018, pp. 1231–45, doi:10.1534/genetics.117.300638.","chicago":"Ringbauer, Harald, Alexander Kolesnikov, David Field, and Nicholas H Barton. “Estimating Barriers to Gene Flow from Distorted Isolation-by-Distance Patterns.” Genetics. Genetics Society of America, 2018. https://doi.org/10.1534/genetics.117.300638.","ama":"Ringbauer H, Kolesnikov A, Field D, Barton NH. Estimating barriers to gene flow from distorted isolation-by-distance patterns. Genetics. 2018;208(3):1231-1245. doi:10.1534/genetics.117.300638","ieee":"H. Ringbauer, A. Kolesnikov, D. Field, and N. H. Barton, “Estimating barriers to gene flow from distorted isolation-by-distance patterns,” Genetics, vol. 208, no. 3. Genetics Society of America, pp. 1231–1245, 2018.","apa":"Ringbauer, H., Kolesnikov, A., Field, D., & Barton, N. H. (2018). Estimating barriers to gene flow from distorted isolation-by-distance patterns. Genetics. Genetics Society of America. https://doi.org/10.1534/genetics.117.300638","ista":"Ringbauer H, Kolesnikov A, Field D, Barton NH. 2018. Estimating barriers to gene flow from distorted isolation-by-distance patterns. Genetics. 208(3), 1231–1245."},"publication":"Genetics","page":"1231-1245","date_published":"2018-03-01T00:00:00Z","scopus_import":"1","article_processing_charge":"No","day":"01","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"563","intvolume":" 208","title":"Estimating barriers to gene flow from distorted isolation-by-distance patterns","status":"public","oa_version":"Preprint","type":"journal_article","issue":"3","abstract":[{"lang":"eng","text":"In continuous populations with local migration, nearby pairs of individuals have on average more similar genotypes\r\nthan geographically well separated pairs. A barrier to gene flow distorts this classical pattern of isolation by distance. Genetic similarity is decreased for sample pairs on different sides of the barrier and increased for pairs on the same side near the barrier. Here, we introduce an inference scheme that utilizes this signal to detect and estimate the strength of a linear barrier to gene flow in two-dimensions. We use a diffusion approximation to model the effects of a barrier on the geographical spread of ancestry backwards in time. This approach allows us to calculate the chance of recent coalescence and probability of identity by descent. We introduce an inference scheme that fits these theoretical results to the geographical covariance structure of bialleleic genetic markers. It can estimate the strength of the barrier as well as several demographic parameters. We investigate the power of our inference scheme to detect barriers by applying it to a wide range of simulated data. We also showcase an example application to a Antirrhinum majus (snapdragon) flower color hybrid zone, where we do not detect any signal of a strong genome wide barrier to gene flow."}],"oa":1,"main_file_link":[{"url":"https://www.biorxiv.org/content/10.1101/205484v1","open_access":"1"}],"external_id":{"isi":["000426219600025"]},"quality_controlled":"1","isi":1,"doi":"10.1534/genetics.117.300638","language":[{"iso":"eng"}],"month":"03","year":"2018","department":[{"_id":"NiBa"},{"_id":"ChLa"}],"publisher":"Genetics Society of America","publication_status":"published","related_material":{"record":[{"relation":"dissertation_contains","status":"public","id":"200"}]},"author":[{"full_name":"Ringbauer, Harald","id":"417FCFF4-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-4884-9682","first_name":"Harald","last_name":"Ringbauer"},{"last_name":"Kolesnikov","first_name":"Alexander","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","full_name":"Kolesnikov, Alexander"},{"full_name":"Field, David","last_name":"Field","first_name":"David"},{"full_name":"Barton, Nicholas H","last_name":"Barton","first_name":"Nicholas H","orcid":"0000-0002-8548-5240","id":"4880FE40-F248-11E8-B48F-1D18A9856A87"}],"volume":208,"date_created":"2018-12-11T11:47:12Z","date_updated":"2023-09-11T13:42:38Z","publist_id":"7251"},{"status":"public","ddc":["000"],"title":"Guest editors' introduction to the special section on learning with Shared information for computer vision and multimedia analysis","intvolume":" 40","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"321","oa_version":"Published Version","file":[{"creator":"dernst","content_type":"application/pdf","file_size":141724,"file_name":"2018_IEEE_Darrell.pdf","access_level":"open_access","date_updated":"2020-07-14T12:46:03Z","date_created":"2020-05-14T12:50:48Z","checksum":"b19c75da06faf3291a3ca47dfa50ef63","file_id":"7835","relation":"main_file"}],"type":"journal_article","abstract":[{"text":"The twelve papers in this special section focus on learning systems with shared information for computer vision and multimedia communication analysis. In the real world, a realistic setting for computer vision or multimedia recognition problems is that we have some classes containing lots of training data and many classes containing a small amount of training data. Therefore, how to use frequent classes to help learning rare classes for which it is harder to collect the training data is an open question. Learning with shared information is an emerging topic in machine learning, computer vision and multimedia analysis. There are different levels of components that can be shared during concept modeling and machine learning stages, such as sharing generic object parts, sharing attributes, sharing transformations, sharing regularization parameters and sharing training examples, etc. Regarding the specific methods, multi-task learning, transfer learning and deep learning can be seen as using different strategies to share information. These learning with shared information methods are very effective in solving real-world large-scale problems.","lang":"eng"}],"issue":"5","article_type":"original","page":"1029 - 1031","publication":"IEEE Transactions on Pattern Analysis and Machine Intelligence","citation":{"short":"T. Darrell, C. Lampert, N. Sebe, Y. Wu, Y. Yan, IEEE Transactions on Pattern Analysis and Machine Intelligence 40 (2018) 1029–1031.","mla":"Darrell, Trevor, et al. “Guest Editors’ Introduction to the Special Section on Learning with Shared Information for Computer Vision and Multimedia Analysis.” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no. 5, IEEE, 2018, pp. 1029–31, doi:10.1109/TPAMI.2018.2804998.","chicago":"Darrell, Trevor, Christoph Lampert, Nico Sebe, Ying Wu, and Yan Yan. “Guest Editors’ Introduction to the Special Section on Learning with Shared Information for Computer Vision and Multimedia Analysis.” IEEE Transactions on Pattern Analysis and Machine Intelligence. IEEE, 2018. https://doi.org/10.1109/TPAMI.2018.2804998.","ama":"Darrell T, Lampert C, Sebe N, Wu Y, Yan Y. Guest editors’ introduction to the special section on learning with Shared information for computer vision and multimedia analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2018;40(5):1029-1031. doi:10.1109/TPAMI.2018.2804998","ieee":"T. Darrell, C. Lampert, N. Sebe, Y. Wu, and Y. Yan, “Guest editors’ introduction to the special section on learning with Shared information for computer vision and multimedia analysis,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no. 5. IEEE, pp. 1029–1031, 2018.","apa":"Darrell, T., Lampert, C., Sebe, N., Wu, Y., & Yan, Y. (2018). Guest editors’ introduction to the special section on learning with Shared information for computer vision and multimedia analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence. IEEE. https://doi.org/10.1109/TPAMI.2018.2804998","ista":"Darrell T, Lampert C, Sebe N, Wu Y, Yan Y. 2018. Guest editors’ introduction to the special section on learning with Shared information for computer vision and multimedia analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence. 40(5), 1029–1031."},"date_published":"2018-05-01T00:00:00Z","scopus_import":"1","day":"01","has_accepted_license":"1","article_processing_charge":"No","publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"IEEE","year":"2018","date_updated":"2023-09-11T14:07:54Z","date_created":"2018-12-11T11:45:48Z","volume":40,"author":[{"full_name":"Darrell, Trevor","last_name":"Darrell","first_name":"Trevor"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"},{"last_name":"Sebe","first_name":"Nico","full_name":"Sebe, Nico"},{"full_name":"Wu, Ying","first_name":"Ying","last_name":"Wu"},{"last_name":"Yan","first_name":"Yan","full_name":"Yan, Yan"}],"file_date_updated":"2020-07-14T12:46:03Z","publist_id":"7544","quality_controlled":"1","isi":1,"external_id":{"isi":["000428901200001"]},"oa":1,"language":[{"iso":"eng"}],"doi":"10.1109/TPAMI.2018.2804998","month":"05"},{"external_id":{"arxiv":["1712.08087"],"isi":["000457843609036"]},"main_file_link":[{"open_access":"1","url":" https://doi.org/10.48550/arXiv.1712.08087"}],"oa":1,"quality_controlled":"1","isi":1,"doi":"10.1109/cvpr.2018.00956","conference":{"name":"CVF: Conference on Computer Vision and Pattern Recognition","location":"Salt Lake City, UT, United States","start_date":"2018-06-18","end_date":"2018-06-23"},"language":[{"iso":"eng"}],"publication_identifier":{"eissn":["2575-7075"],"isbn":["9781538664209"]},"month":"12","year":"2018","department":[{"_id":"ChLa"}],"publisher":"IEEE","publication_status":"published","author":[{"first_name":"Jasper","last_name":"Uijlings","full_name":"Uijlings, Jasper"},{"full_name":"Konyushkova, Ksenia","first_name":"Ksenia","last_name":"Konyushkova"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"},{"full_name":"Ferrari, Vittorio","first_name":"Vittorio","last_name":"Ferrari"}],"date_created":"2022-03-18T12:45:09Z","date_updated":"2023-09-19T15:11:49Z","citation":{"ama":"Uijlings J, Konyushkova K, Lampert C, Ferrari V. Learning intelligent dialogs for bounding box annotation. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE; 2018:9175-9184. doi:10.1109/cvpr.2018.00956","ieee":"J. Uijlings, K. Konyushkova, C. Lampert, and V. Ferrari, “Learning intelligent dialogs for bounding box annotation,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, United States, 2018, pp. 9175–9184.","apa":"Uijlings, J., Konyushkova, K., Lampert, C., & Ferrari, V. (2018). Learning intelligent dialogs for bounding box annotation. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 9175–9184). Salt Lake City, UT, United States: IEEE. https://doi.org/10.1109/cvpr.2018.00956","ista":"Uijlings J, Konyushkova K, Lampert C, Ferrari V. 2018. Learning intelligent dialogs for bounding box annotation. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVF: Conference on Computer Vision and Pattern Recognition, 9175–9184.","short":"J. Uijlings, K. Konyushkova, C. Lampert, V. Ferrari, in:, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2018, pp. 9175–9184.","mla":"Uijlings, Jasper, et al. “Learning Intelligent Dialogs for Bounding Box Annotation.” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2018, pp. 9175–84, doi:10.1109/cvpr.2018.00956.","chicago":"Uijlings, Jasper, Ksenia Konyushkova, Christoph Lampert, and Vittorio Ferrari. “Learning Intelligent Dialogs for Bounding Box Annotation.” In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 9175–84. IEEE, 2018. https://doi.org/10.1109/cvpr.2018.00956."},"publication":"2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition","page":"9175-9184","date_published":"2018-12-17T00:00:00Z","scopus_import":"1","article_processing_charge":"No","day":"17","_id":"10882","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","status":"public","title":"Learning intelligent dialogs for bounding box annotation","oa_version":"Preprint","type":"conference","abstract":[{"lang":"eng","text":"We introduce Intelligent Annotation Dialogs for bounding box annotation. We train an agent to automatically choose a sequence of actions for a human annotator to produce a bounding box in a minimal amount of time. Specifically, we consider two actions: box verification [34], where the annotator verifies a box generated by an object detector, and manual box drawing. We explore two kinds of agents, one based on predicting the probability that a box will be positively verified, and the other based on reinforcement learning. We demonstrate that (1) our agents are able to learn efficient annotation strategies in several scenarios, automatically adapting to the image difficulty, the desired quality of the boxes, and the detector strength; (2) in all scenarios the resulting annotation dialogs speed up annotation compared to manual box drawing alone and box verification alone, while also outperforming any fixed combination of verification and drawing in most scenarios; (3) in a realistic scenario where the detector is iteratively re-trained, our agents evolve a series of strategies that reflect the shifting trade-off between verification and drawing as the detector grows stronger."}]},{"department":[{"_id":"ChLa"}],"publisher":"ML Research Press","publication_status":"published","year":"2018","volume":80,"date_updated":"2023-10-17T09:50:53Z","date_created":"2019-02-14T15:21:07Z","related_material":{"link":[{"description":"News on IST Homepage","relation":"press_release","url":"https://ist.ac.at/en/news/first-machine-learning-method-capable-of-accurate-extrapolation/"}]},"author":[{"full_name":"Sahoo, Subham","first_name":"Subham","last_name":"Sahoo"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"},{"last_name":"Martius","first_name":"Georg S","id":"3A276B68-F248-11E8-B48F-1D18A9856A87","full_name":"Martius, Georg S"}],"ec_funded":1,"project":[{"name":"International IST Postdoc Fellowship Programme","call_identifier":"FP7","grant_number":"291734","_id":"25681D80-B435-11E9-9278-68D0E5697425"}],"quality_controlled":"1","isi":1,"oa":1,"main_file_link":[{"url":"https://arxiv.org/abs/1806.07259","open_access":"1"}],"external_id":{"arxiv":["1806.07259"],"isi":["000683379204058"]},"language":[{"iso":"eng"}],"conference":{"end_date":"2018-07-15","location":"Stockholm, Sweden","start_date":"2018-07-10","name":"ICML: International Conference on Machine Learning"},"month":"02","intvolume":" 80","title":"Learning equations for extrapolation and control","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"6012","oa_version":"Preprint","type":"conference","abstract":[{"lang":"eng","text":"We present an approach to identify concise equations from data using a shallow neural network approach. In contrast to ordinary black-box regression, this approach allows understanding functional relations and generalizing them from observed data to unseen parts of the parameter space. We show how to extend the class of learnable equations for a recently proposed equation learning network to include divisions, and we improve the learning and model selection strategy to be useful for challenging real-world data. For systems governed by analytical expressions, our method can in many cases identify the true underlying equation and extrapolate to unseen domains. We demonstrate its effectiveness by experiments on a cart-pendulum system, where only 2 random rollouts are required to learn the forward dynamics and successfully achieve the swing-up task."}],"page":"4442-4450","citation":{"mla":"Sahoo, Subham, et al. “Learning Equations for Extrapolation and Control.” Proceedings of the 35th International Conference on Machine Learning, vol. 80, ML Research Press, 2018, pp. 4442–50.","short":"S. Sahoo, C. Lampert, G.S. Martius, in:, Proceedings of the 35th International Conference on Machine Learning, ML Research Press, 2018, pp. 4442–4450.","chicago":"Sahoo, Subham, Christoph Lampert, and Georg S Martius. “Learning Equations for Extrapolation and Control.” In Proceedings of the 35th International Conference on Machine Learning, 80:4442–50. ML Research Press, 2018.","ama":"Sahoo S, Lampert C, Martius GS. Learning equations for extrapolation and control. In: Proceedings of the 35th International Conference on Machine Learning. Vol 80. ML Research Press; 2018:4442-4450.","ista":"Sahoo S, Lampert C, Martius GS. 2018. Learning equations for extrapolation and control. Proceedings of the 35th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 80, 4442–4450.","apa":"Sahoo, S., Lampert, C., & Martius, G. S. (2018). Learning equations for extrapolation and control. In Proceedings of the 35th International Conference on Machine Learning (Vol. 80, pp. 4442–4450). Stockholm, Sweden: ML Research Press.","ieee":"S. Sahoo, C. Lampert, and G. S. Martius, “Learning equations for extrapolation and control,” in Proceedings of the 35th International Conference on Machine Learning, Stockholm, Sweden, 2018, vol. 80, pp. 4442–4450."},"publication":"Proceedings of the 35th International Conference on Machine Learning","date_published":"2018-02-01T00:00:00Z","scopus_import":"1","article_processing_charge":"No","day":"01"},{"day":"01","article_processing_charge":"No","scopus_import":"1","date_published":"2018-02-01T00:00:00Z","page":"2815-2824","publication":"Proceedings of the 35 th International Conference on Machine Learning","citation":{"ama":"Kuzborskij I, Lampert C. Data-dependent stability of stochastic gradient descent. In: Proceedings of the 35 Th International Conference on Machine Learning. Vol 80. ML Research Press; 2018:2815-2824.","apa":"Kuzborskij, I., & Lampert, C. (2018). Data-dependent stability of stochastic gradient descent. In Proceedings of the 35 th International Conference on Machine Learning (Vol. 80, pp. 2815–2824). Stockholm, Sweden: ML Research Press.","ieee":"I. Kuzborskij and C. Lampert, “Data-dependent stability of stochastic gradient descent,” in Proceedings of the 35 th International Conference on Machine Learning, Stockholm, Sweden, 2018, vol. 80, pp. 2815–2824.","ista":"Kuzborskij I, Lampert C. 2018. Data-dependent stability of stochastic gradient descent. Proceedings of the 35 th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 80, 2815–2824.","short":"I. Kuzborskij, C. Lampert, in:, Proceedings of the 35 Th International Conference on Machine Learning, ML Research Press, 2018, pp. 2815–2824.","mla":"Kuzborskij, Ilja, and Christoph Lampert. “Data-Dependent Stability of Stochastic Gradient Descent.” Proceedings of the 35 Th International Conference on Machine Learning, vol. 80, ML Research Press, 2018, pp. 2815–24.","chicago":"Kuzborskij, Ilja, and Christoph Lampert. “Data-Dependent Stability of Stochastic Gradient Descent.” In Proceedings of the 35 Th International Conference on Machine Learning, 80:2815–24. ML Research Press, 2018."},"abstract":[{"lang":"eng","text":"We establish a data-dependent notion of algorithmic stability for Stochastic Gradient Descent (SGD), and employ it to develop novel generalization bounds. This is in contrast to previous distribution-free algorithmic stability results for SGD which depend on the worst-case constants. By virtue of the data-dependent argument, our bounds provide new insights into learning with SGD on convex and non-convex problems. In the convex case, we show that the bound on the generalization error depends on the risk at the initialization point. In the non-convex case, we prove that the expected curvature of the objective function around the initialization point has crucial influence on the generalization error. In both cases, our results suggest a simple data-driven strategy to stabilize SGD by pre-screening its initialization. As a corollary, our results allow us to show optimistic generalization bounds that exhibit fast convergence rates for SGD subject to a vanishing empirical risk and low noise of stochastic gradient. "}],"type":"conference","oa_version":"Preprint","status":"public","title":"Data-dependent stability of stochastic gradient descent","intvolume":" 80","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"6011","month":"02","language":[{"iso":"eng"}],"conference":{"end_date":"2018-07-15","start_date":"2018-07-10","location":"Stockholm, Sweden","name":"ICML: International Conference on Machine Learning"},"quality_controlled":"1","isi":1,"project":[{"_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036","name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7"}],"external_id":{"isi":["000683379202095"],"arxiv":["1703.01678"]},"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1703.01678"}],"oa":1,"ec_funded":1,"date_updated":"2023-10-17T09:51:13Z","date_created":"2019-02-14T14:51:57Z","volume":80,"author":[{"first_name":"Ilja","last_name":"Kuzborskij","full_name":"Kuzborskij, Ilja"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"ML Research Press","year":"2018"},{"type":"conference","abstract":[{"text":"Distributed training of massive machine learning models, in particular deep neural networks, via Stochastic Gradient Descent (SGD) is becoming commonplace. Several families of communication-reduction methods, such as quantization, large-batch methods, and gradient sparsification, have been proposed. To date, gradient sparsification methods--where each node sorts gradients by magnitude, and only communicates a subset of the components, accumulating the rest locally--are known to yield some of the largest practical gains. Such methods can reduce the amount of communication per step by up to \\emph{three orders of magnitude}, while preserving model accuracy. Yet, this family of methods currently has no theoretical justification. This is the question we address in this paper. We prove that, under analytic assumptions, sparsifying gradients by magnitude with local error correction provides convergence guarantees, for both convex and non-convex smooth objectives, for data-parallel SGD. The main insight is that sparsification methods implicitly maintain bounds on the maximum impact of stale updates, thanks to selection by magnitude. Our analysis and empirical validation also reveal that these methods do require analytical conditions to converge well, justifying existing heuristics.","lang":"eng"}],"_id":"6589","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","title":"The convergence of sparsified gradient methods","status":"public","oa_version":"Preprint","scopus_import":"1","day":"01","article_processing_charge":"No","publication":"Advances in Neural Information Processing Systems 31","citation":{"apa":"Alistarh, D.-A., Hoefler, T., Johansson, M., Konstantinov, N. H., Khirirat, S., & Renggli, C. (2018). The convergence of sparsified gradient methods. In Advances in Neural Information Processing Systems 31 (Vol. Volume 2018, pp. 5973–5983). Montreal, Canada: Neural Information Processing Systems Foundation.","ieee":"D.-A. Alistarh, T. Hoefler, M. Johansson, N. H. Konstantinov, S. Khirirat, and C. Renggli, “The convergence of sparsified gradient methods,” in Advances in Neural Information Processing Systems 31, Montreal, Canada, 2018, vol. Volume 2018, pp. 5973–5983.","ista":"Alistarh D-A, Hoefler T, Johansson M, Konstantinov NH, Khirirat S, Renggli C. 2018. The convergence of sparsified gradient methods. Advances in Neural Information Processing Systems 31. NeurIPS: Conference on Neural Information Processing Systems vol. Volume 2018, 5973–5983.","ama":"Alistarh D-A, Hoefler T, Johansson M, Konstantinov NH, Khirirat S, Renggli C. The convergence of sparsified gradient methods. In: Advances in Neural Information Processing Systems 31. Vol Volume 2018. Neural Information Processing Systems Foundation; 2018:5973-5983.","chicago":"Alistarh, Dan-Adrian, Torsten Hoefler, Mikael Johansson, Nikola H Konstantinov, Sarit Khirirat, and Cedric Renggli. “The Convergence of Sparsified Gradient Methods.” In Advances in Neural Information Processing Systems 31, Volume 2018:5973–83. Neural Information Processing Systems Foundation, 2018.","short":"D.-A. Alistarh, T. Hoefler, M. Johansson, N.H. Konstantinov, S. Khirirat, C. Renggli, in:, Advances in Neural Information Processing Systems 31, Neural Information Processing Systems Foundation, 2018, pp. 5973–5983.","mla":"Alistarh, Dan-Adrian, et al. “The Convergence of Sparsified Gradient Methods.” Advances in Neural Information Processing Systems 31, vol. Volume 2018, Neural Information Processing Systems Foundation, 2018, pp. 5973–83."},"page":"5973-5983","date_published":"2018-12-01T00:00:00Z","ec_funded":1,"year":"2018","publication_status":"published","department":[{"_id":"DaAl"},{"_id":"ChLa"}],"publisher":"Neural Information Processing Systems Foundation","author":[{"last_name":"Alistarh","first_name":"Dan-Adrian","orcid":"0000-0003-3650-940X","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","full_name":"Alistarh, Dan-Adrian"},{"full_name":"Hoefler, Torsten","last_name":"Hoefler","first_name":"Torsten"},{"full_name":"Johansson, Mikael","first_name":"Mikael","last_name":"Johansson"},{"last_name":"Konstantinov","first_name":"Nikola H","id":"4B9D76E4-F248-11E8-B48F-1D18A9856A87","full_name":"Konstantinov, Nikola H"},{"first_name":"Sarit","last_name":"Khirirat","full_name":"Khirirat, Sarit"},{"first_name":"Cedric","last_name":"Renggli","full_name":"Renggli, Cedric"}],"date_created":"2019-06-27T09:32:55Z","date_updated":"2023-10-17T11:47:20Z","volume":"Volume 2018","month":"12","oa":1,"main_file_link":[{"url":"https://arxiv.org/abs/1809.10505","open_access":"1"}],"external_id":{"arxiv":["1809.10505"],"isi":["000461852000047"]},"isi":1,"quality_controlled":"1","project":[{"name":"International IST Doctoral Program","call_identifier":"H2020","_id":"2564DBCA-B435-11E9-9278-68D0E5697425","grant_number":"665385"}],"conference":{"end_date":"2018-12-08","location":"Montreal, Canada","start_date":"2018-12-02","name":"NeurIPS: Conference on Neural Information Processing Systems"},"language":[{"iso":"eng"}]},{"datarep_id":"98","type":"research_data","abstract":[{"text":"This package contains data for the publication \"Nonlinear decoding of a complex movie from the mammalian retina\" by Deny S. et al, PLOS Comput Biol (2018). \r\n\r\nThe data consists of\r\n(i) 91 spike sorted, isolated rat retinal ganglion cells that pass stability and quality criteria, recorded on the multi-electrode array, in response to the presentation of the complex movie with many randomly moving dark discs. The responses are represented as 648000 x 91 binary matrix, where the first index indicates the timebin of duration 12.5 ms, and the second index the neural identity. The matrix entry is 0/1 if the neuron didn't/did spike in the particular time bin.\r\n(ii) README file and a graphical illustration of the structure of the experiment, specifying how the 648000 timebins are split into epochs where 1, 2, 4, or 10 discs were displayed, and which stimulus segments are exact repeats or unique ball trajectories.\r\n(iii) a 648000 x 400 matrix of luminance traces for each of the 20 x 20 positions (\"sites\") in the movie frame, with time that is locked to the recorded raster. The luminance traces are produced as described in the manuscript by filtering the raw disc movie with a small gaussian spatial kernel. ","lang":"eng"}],"file_date_updated":"2020-07-14T12:47:07Z","license":"https://creativecommons.org/publicdomain/zero/1.0/","_id":"5584","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2018","ddc":["570"],"title":"Nonlinear decoding of a complex movie from the mammalian retina","status":"public","publisher":"Institute of Science and Technology Austria","department":[{"_id":"ChLa"},{"_id":"GaTk"}],"author":[{"full_name":"Deny, Stephane","first_name":"Stephane","last_name":"Deny"},{"last_name":"Marre","first_name":"Olivier","full_name":"Marre, Olivier"},{"last_name":"Botella-Soler","first_name":"Vicente","full_name":"Botella-Soler, Vicente"},{"id":"3A276B68-F248-11E8-B48F-1D18A9856A87","last_name":"Martius","first_name":"Georg S","full_name":"Martius, Georg S"},{"full_name":"Tkacik, Gasper","id":"3D494DCA-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-6699-1455","first_name":"Gasper","last_name":"Tkacik"}],"related_material":{"record":[{"id":"292","status":"public","relation":"used_in_publication"}]},"date_created":"2018-12-12T12:31:39Z","date_updated":"2024-02-21T13:45:26Z","oa_version":"Published Version","file":[{"checksum":"6808748837b9afbbbabc2a356ca2b88a","date_created":"2018-12-12T13:02:24Z","date_updated":"2020-07-14T12:47:07Z","relation":"main_file","file_id":"5590","content_type":"application/octet-stream","file_size":1142543971,"creator":"system","access_level":"open_access","file_name":"IST-2018-98-v1+1_BBalls_area2_tile2_20x20.mat"},{"file_name":"IST-2018-98-v1+2_ExperimentStructure.pdf","access_level":"open_access","creator":"system","file_size":702336,"content_type":"application/pdf","file_id":"5591","relation":"main_file","date_updated":"2020-07-14T12:47:07Z","date_created":"2018-12-12T13:02:25Z","checksum":"d6d6cd07743038fe3a12352983fcf9dd"},{"date_created":"2018-12-12T13:02:26Z","date_updated":"2020-07-14T12:47:07Z","checksum":"0c9cfb4dab35bb3dc25a04395600b1c8","file_id":"5592","relation":"main_file","creator":"system","file_size":432,"content_type":"application/octet-stream","file_name":"IST-2018-98-v1+3_GoodLocations_area2_20x20.mat","access_level":"open_access"},{"file_id":"5593","relation":"main_file","checksum":"2a83b011012e21e934b4596285b1a183","date_updated":"2020-07-14T12:47:07Z","date_created":"2018-12-12T13:02:26Z","access_level":"open_access","file_name":"IST-2018-98-v1+4_README.txt","creator":"system","file_size":986,"content_type":"text/plain"}],"keyword":["retina","decoding","regression","neural networks","complex stimulus"],"month":"03","day":"29","article_processing_charge":"No","has_accepted_license":"1","citation":{"chicago":"Deny, Stephane, Olivier Marre, Vicente Botella-Soler, Georg S Martius, and Gašper Tkačik. “Nonlinear Decoding of a Complex Movie from the Mammalian Retina.” Institute of Science and Technology Austria, 2018. https://doi.org/10.15479/AT:ISTA:98.","short":"S. Deny, O. Marre, V. Botella-Soler, G.S. Martius, G. Tkačik, (2018).","mla":"Deny, Stephane, et al. Nonlinear Decoding of a Complex Movie from the Mammalian Retina. Institute of Science and Technology Austria, 2018, doi:10.15479/AT:ISTA:98.","apa":"Deny, S., Marre, O., Botella-Soler, V., Martius, G. S., & Tkačik, G. (2018). Nonlinear decoding of a complex movie from the mammalian retina. Institute of Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:98","ieee":"S. Deny, O. Marre, V. Botella-Soler, G. S. Martius, and G. Tkačik, “Nonlinear decoding of a complex movie from the mammalian retina.” Institute of Science and Technology Austria, 2018.","ista":"Deny S, Marre O, Botella-Soler V, Martius GS, Tkačik G. 2018. Nonlinear decoding of a complex movie from the mammalian retina, Institute of Science and Technology Austria, 10.15479/AT:ISTA:98.","ama":"Deny S, Marre O, Botella-Soler V, Martius GS, Tkačik G. Nonlinear decoding of a complex movie from the mammalian retina. 2018. doi:10.15479/AT:ISTA:98"},"tmp":{"short":"CC0 (1.0)","image":"/images/cc_0.png","legal_code_url":"https://creativecommons.org/publicdomain/zero/1.0/legalcode","name":"Creative Commons Public Domain Dedication (CC0 1.0)"},"oa":1,"project":[{"_id":"254D1A94-B435-11E9-9278-68D0E5697425","grant_number":"P 25651-N26","call_identifier":"FWF","name":"Sensitivity to higher-order statistics in natural scenes"}],"doi":"10.15479/AT:ISTA:98","date_published":"2018-03-29T00:00:00Z"},{"doi":"10.1109/DEVLRN.2016.7846789","date_published":"2017-02-07T00:00:00Z","conference":{"name":"ICDL EpiRob: International Conference on Development and Learning and Epigenetic Robotics ","start_date":"2016-09-19","location":"Cergy-Pontoise, France","end_date":"2016-09-22"},"language":[{"iso":"eng"}],"citation":{"ieee":"R. Der and G. S. Martius, “Dynamical self consistency leads to behavioral development and emergent social interactions in robots,” presented at the ICDL EpiRob: International Conference on Development and Learning and Epigenetic Robotics , Cergy-Pontoise, France, 2017.","apa":"Der, R., & Martius, G. S. (2017). Dynamical self consistency leads to behavioral development and emergent social interactions in robots. Presented at the ICDL EpiRob: International Conference on Development and Learning and Epigenetic Robotics , Cergy-Pontoise, France: IEEE. https://doi.org/10.1109/DEVLRN.2016.7846789","ista":"Der R, Martius GS. 2017. Dynamical self consistency leads to behavioral development and emergent social interactions in robots. ICDL EpiRob: International Conference on Development and Learning and Epigenetic Robotics , 7846789.","ama":"Der R, Martius GS. Dynamical self consistency leads to behavioral development and emergent social interactions in robots. In: IEEE; 2017. doi:10.1109/DEVLRN.2016.7846789","chicago":"Der, Ralf, and Georg S Martius. “Dynamical Self Consistency Leads to Behavioral Development and Emergent Social Interactions in Robots.” IEEE, 2017. https://doi.org/10.1109/DEVLRN.2016.7846789.","short":"R. Der, G.S. Martius, in:, IEEE, 2017.","mla":"Der, Ralf, and Georg S. Martius. Dynamical Self Consistency Leads to Behavioral Development and Emergent Social Interactions in Robots. 7846789, IEEE, 2017, doi:10.1109/DEVLRN.2016.7846789."},"quality_controlled":"1","publication_identifier":{"isbn":["978-150905069-7"]},"day":"07","month":"02","scopus_import":1,"author":[{"full_name":"Der, Ralf","first_name":"Ralf","last_name":"Der"},{"full_name":"Martius, Georg S","id":"3A276B68-F248-11E8-B48F-1D18A9856A87","first_name":"Georg S","last_name":"Martius"}],"oa_version":"None","date_updated":"2021-01-12T08:07:51Z","date_created":"2018-12-11T11:47:43Z","_id":"652","year":"2017","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","publisher":"IEEE","department":[{"_id":"ChLa"},{"_id":"GaTk"}],"title":"Dynamical self consistency leads to behavioral development and emergent social interactions in robots","publication_status":"published","status":"public","publist_id":"7100","abstract":[{"lang":"eng","text":"We present an approach that enables robots to self-organize their sensorimotor behavior from scratch without providing specific information about neither the robot nor its environment. This is achieved by a simple neural control law that increases the consistency between external sensor dynamics and internal neural dynamics of the utterly simple controller. In this way, the embodiment and the agent-environment coupling are the only source of individual development. We show how an anthropomorphic tendon driven arm-shoulder system develops different behaviors depending on that coupling. For instance: Given a bottle half-filled with water, the arm starts to shake it, driven by the physical response of the water. When attaching a brush, the arm can be manipulated into wiping a table, and when connected to a revolvable wheel it finds out how to rotate it. Thus, the robot may be said to discover the affordances of the world. When allowing two (simulated) humanoid robots to interact physically, they engage into a joint behavior development leading to, for instance, spontaneous cooperation. More social effects are observed if the robots can visually perceive each other. Although, as an observer, it is tempting to attribute an apparent intentionality, there is nothing of the kind put in. As a conclusion, we argue that emergent behavior may be much less rooted in explicit intentions, internal motivations, or specific reward systems than is commonly believed."}],"type":"conference","article_number":"7846789"},{"publication_identifier":{"issn":["16625218"]},"month":"03","language":[{"iso":"eng"}],"doi":"10.3389/fnbot.2017.00008","project":[{"_id":"25681D80-B435-11E9-9278-68D0E5697425","grant_number":"291734","name":"International IST Postdoc Fellowship Programme","call_identifier":"FP7"}],"quality_controlled":"1","tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"oa":1,"ec_funded":1,"publist_id":"7078","file_date_updated":"2020-07-14T12:47:33Z","article_number":"00008","volume":11,"date_created":"2018-12-11T11:47:45Z","date_updated":"2021-01-12T08:08:04Z","author":[{"full_name":"Der, Ralf","last_name":"Der","first_name":"Ralf"},{"id":"3A276B68-F248-11E8-B48F-1D18A9856A87","first_name":"Georg S","last_name":"Martius","full_name":"Martius, Georg S"}],"publisher":"Frontiers Research Foundation","department":[{"_id":"ChLa"},{"_id":"GaTk"}],"publication_status":"published","year":"2017","article_processing_charge":"Yes","has_accepted_license":"1","day":"16","scopus_import":1,"date_published":"2017-03-16T00:00:00Z","citation":{"ista":"Der R, Martius GS. 2017. Self organized behavior generation for musculoskeletal robots. Frontiers in Neurorobotics. 11(MAR), 00008.","ieee":"R. Der and G. S. Martius, “Self organized behavior generation for musculoskeletal robots,” Frontiers in Neurorobotics, vol. 11, no. MAR. Frontiers Research Foundation, 2017.","apa":"Der, R., & Martius, G. S. (2017). Self organized behavior generation for musculoskeletal robots. Frontiers in Neurorobotics. Frontiers Research Foundation. https://doi.org/10.3389/fnbot.2017.00008","ama":"Der R, Martius GS. Self organized behavior generation for musculoskeletal robots. Frontiers in Neurorobotics. 2017;11(MAR). doi:10.3389/fnbot.2017.00008","chicago":"Der, Ralf, and Georg S Martius. “Self Organized Behavior Generation for Musculoskeletal Robots.” Frontiers in Neurorobotics. Frontiers Research Foundation, 2017. https://doi.org/10.3389/fnbot.2017.00008.","mla":"Der, Ralf, and Georg S. Martius. “Self Organized Behavior Generation for Musculoskeletal Robots.” Frontiers in Neurorobotics, vol. 11, no. MAR, 00008, Frontiers Research Foundation, 2017, doi:10.3389/fnbot.2017.00008.","short":"R. Der, G.S. Martius, Frontiers in Neurorobotics 11 (2017)."},"publication":"Frontiers in Neurorobotics","issue":"MAR","abstract":[{"lang":"eng","text":"With the accelerated development of robot technologies, control becomes one of the central themes of research. In traditional approaches, the controller, by its internal functionality, finds appropriate actions on the basis of specific objectives for the task at hand. While very successful in many applications, self-organized control schemes seem to be favored in large complex systems with unknown dynamics or which are difficult to model. Reasons are the expected scalability, robustness, and resilience of self-organizing systems. The paper presents a self-learning neurocontroller based on extrinsic differential plasticity introduced recently, applying it to an anthropomorphic musculoskeletal robot arm with attached objects of unknown physical dynamics. The central finding of the paper is the following effect: by the mere feedback through the internal dynamics of the object, the robot is learning to relate each of the objects with a very specific sensorimotor pattern. Specifically, an attached pendulum pilots the arm into a circular motion, a half-filled bottle produces axis oriented shaking behavior, a wheel is getting rotated, and wiping patterns emerge automatically in a table-plus-brush setting. By these object-specific dynamical patterns, the robot may be said to recognize the object's identity, or in other words, it discovers dynamical affordances of objects. Furthermore, when including hand coordinates obtained from a camera, a dedicated hand-eye coordination self-organizes spontaneously. These phenomena are discussed from a specific dynamical system perspective. Central is the dedicated working regime at the border to instability with its potentially infinite reservoir of (limit cycle) attractors "waiting" to be excited. Besides converging toward one of these attractors, variate behavior is also arising from a self-induced attractor morphing driven by the learning rule. We claim that experimental investigations with this anthropomorphic, self-learning robot not only generate interesting and potentially useful behaviors, but may also help to better understand what subjective human muscle feelings are, how they can be rooted in sensorimotor patterns, and how these concepts may feed back on robotics."}],"type":"journal_article","oa_version":"Published Version","file":[{"file_id":"5371","relation":"main_file","date_updated":"2020-07-14T12:47:33Z","date_created":"2018-12-12T10:18:49Z","checksum":"b1bc43f96d1df3313c03032c2a46388d","file_name":"IST-2017-903-v1+1_fnbot-11-00008.pdf","access_level":"open_access","creator":"system","content_type":"application/pdf","file_size":8439566}],"pubrep_id":"903","intvolume":" 11","ddc":["006"],"status":"public","title":"Self organized behavior generation for musculoskeletal robots","_id":"658","user_id":"2EBD1598-F248-11E8-B48F-1D18A9856A87"},{"abstract":[{"text":"In classical machine learning, regression is treated as a black box process of identifying a suitable function from a hypothesis set without attempting to gain insight into the mechanism connecting inputs and outputs. In the natural sciences, however, finding an interpretable function for a phenomenon is the prime goal as it allows to understand and generalize results. This paper proposes a novel type of function learning network, called equation learner (EQL), that can learn analytical expressions and is able to extrapolate to unseen domains. It is implemented as an end-to-end differentiable feed-forward network and allows for efficient gradient based training. Due to sparsity regularization concise interpretable expressions can be obtained. Often the true underlying source expression is identified.","lang":"eng"}],"ec_funded":1,"type":"conference","author":[{"first_name":"Georg S","last_name":"Martius","id":"3A276B68-F248-11E8-B48F-1D18A9856A87","full_name":"Martius, Georg S"},{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph"}],"date_created":"2019-09-01T22:01:00Z","date_updated":"2021-01-12T08:09:17Z","oa_version":"Preprint","_id":"6841","year":"2017","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","publication_status":"published","title":"Extrapolation and learning equations","status":"public","publisher":"International Conference on Learning Representations","department":[{"_id":"ChLa"}],"day":"21","month":"02","scopus_import":1,"conference":{"name":"ICLR: International Conference on Learning Representations","end_date":"2017-04-26","start_date":"2017-04-24","location":"Toulon, France"},"date_published":"2017-02-21T00:00:00Z","language":[{"iso":"eng"}],"publication":"5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings","main_file_link":[{"url":"https://arxiv.org/abs/1610.02995","open_access":"1"}],"citation":{"apa":"Martius, G. S., & Lampert, C. (2017). Extrapolation and learning equations. In 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings. Toulon, France: International Conference on Learning Representations.","ieee":"G. S. Martius and C. Lampert, “Extrapolation and learning equations,” in 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings, Toulon, France, 2017.","ista":"Martius GS, Lampert C. 2017. Extrapolation and learning equations. 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings. ICLR: International Conference on Learning Representations.","ama":"Martius GS, Lampert C. Extrapolation and learning equations. In: 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings. International Conference on Learning Representations; 2017.","chicago":"Martius, Georg S, and Christoph Lampert. “Extrapolation and Learning Equations.” In 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings. International Conference on Learning Representations, 2017.","short":"G.S. Martius, C. Lampert, in:, 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings, International Conference on Learning Representations, 2017.","mla":"Martius, Georg S., and Christoph Lampert. “Extrapolation and Learning Equations.” 5th International Conference on Learning Representations, ICLR 2017 - Workshop Track Proceedings, International Conference on Learning Representations, 2017."},"external_id":{"arxiv":["1610.02995"]},"oa":1,"quality_controlled":"1","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}]},{"type":"conference","publist_id":"6906","abstract":[{"lang":"eng","text":"Modern communication technologies allow first responders to contact thousands of potential volunteers simultaneously for support during a crisis or disaster event. However, such volunteer efforts must be well coordinated and monitored, in order to offer an effective relief to the professionals. In this paper we extend earlier work on optimally assigning volunteers to selected landmark locations. In particular, we emphasize the aspect that obtaining good assignments requires not only advanced computational tools, but also a realistic measure of distance between volunteers and landmarks. Specifically, we propose the use of the Open Street Map (OSM) driving distance instead of he previously used flight distance. We find the OSM driving distance to be better aligned with the interests of volunteers and first responders. Furthermore, we show that relying on the flying distance leads to a substantial underestimation of the number of required volunteers, causing negative side effects in case of an actual crisis situation."}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"750","year":"2017","department":[{"_id":"ChLa"}],"publisher":"IEEE","publication_status":"published","status":"public","title":"Optimal geospatial volunteer allocation needs realistic distances","author":[{"first_name":"Jasmin","last_name":"Pielorz","id":"49BC895A-F248-11E8-B48F-1D18A9856A87","full_name":"Pielorz, Jasmin"},{"full_name":"Prandtstetter, Matthias","first_name":"Matthias","last_name":"Prandtstetter"},{"full_name":"Straub, Markus","first_name":"Markus","last_name":"Straub"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"oa_version":"None","date_created":"2018-12-11T11:48:18Z","date_updated":"2021-01-12T08:13:55Z","scopus_import":1,"publication_identifier":{"isbn":["978-153862714-3"]},"day":"01","month":"12","citation":{"ama":"Pielorz J, Prandtstetter M, Straub M, Lampert C. Optimal geospatial volunteer allocation needs realistic distances. In: 2017 IEEE International Conference on Big Data. IEEE; 2017:3760-3763. doi:10.1109/BigData.2017.8258375","apa":"Pielorz, J., Prandtstetter, M., Straub, M., & Lampert, C. (2017). Optimal geospatial volunteer allocation needs realistic distances. In 2017 IEEE International Conference on Big Data (pp. 3760–3763). Boston, MA, United States: IEEE. https://doi.org/10.1109/BigData.2017.8258375","ieee":"J. Pielorz, M. Prandtstetter, M. Straub, and C. Lampert, “Optimal geospatial volunteer allocation needs realistic distances,” in 2017 IEEE International Conference on Big Data, Boston, MA, United States, 2017, pp. 3760–3763.","ista":"Pielorz J, Prandtstetter M, Straub M, Lampert C. 2017. Optimal geospatial volunteer allocation needs realistic distances. 2017 IEEE International Conference on Big Data. Big Data, 3760–3763.","short":"J. Pielorz, M. Prandtstetter, M. Straub, C. Lampert, in:, 2017 IEEE International Conference on Big Data, IEEE, 2017, pp. 3760–3763.","mla":"Pielorz, Jasmin, et al. “Optimal Geospatial Volunteer Allocation Needs Realistic Distances.” 2017 IEEE International Conference on Big Data, IEEE, 2017, pp. 3760–63, doi:10.1109/BigData.2017.8258375.","chicago":"Pielorz, Jasmin, Matthias Prandtstetter, Markus Straub, and Christoph Lampert. “Optimal Geospatial Volunteer Allocation Needs Realistic Distances.” In 2017 IEEE International Conference on Big Data, 3760–63. IEEE, 2017. https://doi.org/10.1109/BigData.2017.8258375."},"publication":"2017 IEEE International Conference on Big Data","page":"3760 - 3763","quality_controlled":"1","doi":"10.1109/BigData.2017.8258375","date_published":"2017-12-01T00:00:00Z","conference":{"name":"Big Data","start_date":"2017-12-11","location":"Boston, MA, United States","end_date":"2017-12-14"},"language":[{"iso":"eng"}]},{"year":"2017","acknowledgement":"We thank Tim Salimans for spotting a mistake in our preliminary arXiv manuscript. This work was funded by the European Research Council under the European Unions Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no 308036.","publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"JMLR","author":[{"full_name":"Kolesnikov, Alexander","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","first_name":"Alexander","last_name":"Kolesnikov"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"date_created":"2018-12-11T11:49:37Z","date_updated":"2023-09-22T09:50:41Z","volume":70,"publist_id":"6398","ec_funded":1,"external_id":{"arxiv":["1612.08185"],"isi":["000683309501102"]},"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1612.08185"}],"oa":1,"quality_controlled":"1","isi":1,"project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"conference":{"start_date":"2017-08-06","location":"Sydney, Australia","end_date":"2017-08-11","name":"ICML: International Conference on Machine Learning"},"language":[{"iso":"eng"}],"month":"08","publication_identifier":{"isbn":["978-151085514-4"]},"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"1000","title":"PixelCNN models with auxiliary variables for natural image modeling","status":"public","intvolume":" 70","oa_version":"Submitted Version","type":"conference","abstract":[{"lang":"eng","text":"We study probabilistic models of natural images and extend the autoregressive family of PixelCNN models by incorporating latent variables. Subsequently, we describe two new generative image models that exploit different image transformations as latent variables: a quantized grayscale view of the image or a multi-resolution image pyramid. The proposed models tackle two known shortcomings of existing PixelCNN models: 1) their tendency to focus on low-level image details, while largely ignoring high-level image information, such as object shapes, and 2) their computationally costly procedure for image sampling. We experimentally demonstrate benefits of our LatentPixelCNN models, in particular showing that they produce much more realistically looking image samples than previous state-of-the-art probabilistic models. "}],"publication":"34th International Conference on Machine Learning","citation":{"ama":"Kolesnikov A, Lampert C. PixelCNN models with auxiliary variables for natural image modeling. In: 34th International Conference on Machine Learning. Vol 70. JMLR; 2017:1905-1914.","ista":"Kolesnikov A, Lampert C. 2017. PixelCNN models with auxiliary variables for natural image modeling. 34th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 70, 1905–1914.","ieee":"A. Kolesnikov and C. Lampert, “PixelCNN models with auxiliary variables for natural image modeling,” in 34th International Conference on Machine Learning, Sydney, Australia, 2017, vol. 70, pp. 1905–1914.","apa":"Kolesnikov, A., & Lampert, C. (2017). PixelCNN models with auxiliary variables for natural image modeling. In 34th International Conference on Machine Learning (Vol. 70, pp. 1905–1914). Sydney, Australia: JMLR.","mla":"Kolesnikov, Alexander, and Christoph Lampert. “PixelCNN Models with Auxiliary Variables for Natural Image Modeling.” 34th International Conference on Machine Learning, vol. 70, JMLR, 2017, pp. 1905–14.","short":"A. Kolesnikov, C. Lampert, in:, 34th International Conference on Machine Learning, JMLR, 2017, pp. 1905–1914.","chicago":"Kolesnikov, Alexander, and Christoph Lampert. “PixelCNN Models with Auxiliary Variables for Natural Image Modeling.” In 34th International Conference on Machine Learning, 70:1905–14. JMLR, 2017."},"page":"1905 - 1914","date_published":"2017-08-01T00:00:00Z","scopus_import":"1","day":"01","article_processing_charge":"No","has_accepted_license":"1"},{"type":"conference","abstract":[{"text":"A major open problem on the road to artificial intelligence is the development of incrementally learning systems that learn about more and more concepts over time from a stream of data. In this work, we introduce a new training strategy, iCaRL, that allows learning in such a class-incremental way: only the training data for a small number of classes has to be present at the same time and new classes can be added progressively. iCaRL learns strong classifiers and a data representation simultaneously. This distinguishes it from earlier works that were fundamentally limited to fixed data representations and therefore incompatible with deep learning architectures. We show by experiments on CIFAR-100 and ImageNet ILSVRC 2012 data that iCaRL can learn many classes incrementally over a long period of time where other strategies quickly fail. ","lang":"eng"}],"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","_id":"998","intvolume":" 2017","status":"public","title":"iCaRL: Incremental classifier and representation learning","oa_version":"Submitted Version","scopus_import":"1","article_processing_charge":"No","day":"14","citation":{"chicago":"Rebuffi, Sylvestre Alvise, Alexander Kolesnikov, Georg Sperl, and Christoph Lampert. “ICaRL: Incremental Classifier and Representation Learning,” 2017:5533–42. IEEE, 2017. https://doi.org/10.1109/CVPR.2017.587.","mla":"Rebuffi, Sylvestre Alvise, et al. ICaRL: Incremental Classifier and Representation Learning. Vol. 2017, IEEE, 2017, pp. 5533–42, doi:10.1109/CVPR.2017.587.","short":"S.A. Rebuffi, A. Kolesnikov, G. Sperl, C. Lampert, in:, IEEE, 2017, pp. 5533–5542.","ista":"Rebuffi SA, Kolesnikov A, Sperl G, Lampert C. 2017. iCaRL: Incremental classifier and representation learning. CVPR: Computer Vision and Pattern Recognition vol. 2017, 5533–5542.","apa":"Rebuffi, S. A., Kolesnikov, A., Sperl, G., & Lampert, C. (2017). iCaRL: Incremental classifier and representation learning (Vol. 2017, pp. 5533–5542). Presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States: IEEE. https://doi.org/10.1109/CVPR.2017.587","ieee":"S. A. Rebuffi, A. Kolesnikov, G. Sperl, and C. Lampert, “iCaRL: Incremental classifier and representation learning,” presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States, 2017, vol. 2017, pp. 5533–5542.","ama":"Rebuffi SA, Kolesnikov A, Sperl G, Lampert C. iCaRL: Incremental classifier and representation learning. In: Vol 2017. IEEE; 2017:5533-5542. doi:10.1109/CVPR.2017.587"},"page":"5533 - 5542","date_published":"2017-04-14T00:00:00Z","ec_funded":1,"publist_id":"6400","year":"2017","publisher":"IEEE","department":[{"_id":"ChLa"},{"_id":"ChWo"}],"publication_status":"published","author":[{"full_name":"Rebuffi, Sylvestre Alvise","last_name":"Rebuffi","first_name":"Sylvestre Alvise"},{"full_name":"Kolesnikov, Alexander","first_name":"Alexander","last_name":"Kolesnikov","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Sperl, Georg","id":"4DD40360-F248-11E8-B48F-1D18A9856A87","first_name":"Georg","last_name":"Sperl"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"volume":2017,"date_updated":"2023-09-22T09:51:58Z","date_created":"2018-12-11T11:49:37Z","publication_identifier":{"isbn":["978-153860457-1"]},"month":"04","oa":1,"main_file_link":[{"url":"https://arxiv.org/abs/1611.07725","open_access":"1"}],"external_id":{"isi":["000418371405066"]},"project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"quality_controlled":"1","isi":1,"doi":"10.1109/CVPR.2017.587","conference":{"name":"CVPR: Computer Vision and Pattern Recognition","start_date":"2017-07-21","location":"Honolulu, HA, United States","end_date":"2017-07-26"},"language":[{"iso":"eng"}]},{"type":"conference","abstract":[{"text":"We develop a probabilistic technique for colorizing grayscale natural images. In light of the intrinsic uncertainty of this task, the proposed probabilistic framework has numerous desirable properties. In particular, our model is able to produce multiple plausible and vivid colorizations for a given grayscale image and is one of the first colorization models to provide a proper stochastic sampling scheme. Moreover, our training procedure is supported by a rigorous theoretical framework that does not require any ad hoc heuristics and allows for efficient modeling and learning of the joint pixel color distribution.We demonstrate strong quantitative and qualitative experimental results on the CIFAR-10 dataset and the challenging ILSVRC 2012 dataset.","lang":"eng"}],"ddc":["000"],"title":"Probabilistic image colorization","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"911","oa_version":"Published Version","file":[{"creator":"dernst","content_type":"application/pdf","file_size":1625363,"access_level":"open_access","file_name":"2017_BMVC_Royer.pdf","success":1,"date_created":"2020-08-10T07:14:33Z","date_updated":"2020-08-10T07:14:33Z","file_id":"8224","relation":"main_file"}],"scopus_import":"1","day":"01","article_processing_charge":"No","has_accepted_license":"1","page":"85.1-85.12","citation":{"mla":"Royer, Amélie, et al. Probabilistic Image Colorization. BMVA Press, 2017, p. 85.1-85.12, doi:10.5244/c.31.85.","short":"A. Royer, A. Kolesnikov, C. Lampert, in:, BMVA Press, 2017, p. 85.1-85.12.","chicago":"Royer, Amélie, Alexander Kolesnikov, and Christoph Lampert. “Probabilistic Image Colorization,” 85.1-85.12. BMVA Press, 2017. https://doi.org/10.5244/c.31.85.","ama":"Royer A, Kolesnikov A, Lampert C. Probabilistic image colorization. In: BMVA Press; 2017:85.1-85.12. doi:10.5244/c.31.85","ista":"Royer A, Kolesnikov A, Lampert C. 2017. Probabilistic image colorization. BMVC: British Machine Vision Conference, 85.1-85.12.","ieee":"A. Royer, A. Kolesnikov, and C. Lampert, “Probabilistic image colorization,” presented at the BMVC: British Machine Vision Conference, London, United Kingdom, 2017, p. 85.1-85.12.","apa":"Royer, A., Kolesnikov, A., & Lampert, C. (2017). Probabilistic image colorization (p. 85.1-85.12). Presented at the BMVC: British Machine Vision Conference, London, United Kingdom: BMVA Press. https://doi.org/10.5244/c.31.85"},"date_published":"2017-09-01T00:00:00Z","file_date_updated":"2020-08-10T07:14:33Z","ec_funded":1,"publist_id":"6532","publication_status":"published","publisher":"BMVA Press","department":[{"_id":"ChLa"}],"year":"2017","date_updated":"2023-10-16T10:04:02Z","date_created":"2018-12-11T11:49:09Z","author":[{"full_name":"Royer, Amélie","last_name":"Royer","first_name":"Amélie","orcid":"0000-0002-8407-0705","id":"3811D890-F248-11E8-B48F-1D18A9856A87"},{"id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","last_name":"Kolesnikov","first_name":"Alexander","full_name":"Kolesnikov, Alexander"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"8390"}]},"month":"09","publication_identifier":{"eisbn":["190172560X"]},"quality_controlled":"1","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"external_id":{"arxiv":["1705.04258"]},"oa":1,"language":[{"iso":"eng"}],"conference":{"name":"BMVC: British Machine Vision Conference","start_date":"2017-09-04","location":"London, United Kingdom","end_date":"2017-09-07"},"doi":"10.5244/c.31.85"},{"alternative_title":["PMLR"],"type":"conference","abstract":[{"lang":"eng","text":"In this work we study the learnability of stochastic processes with respect to the conditional risk, i.e. the existence of a learning algorithm that improves its next-step performance with the amount of observed data. We introduce a notion of pairwise discrepancy between conditional distributions at different times steps and show how certain properties of these discrepancies can be used to construct a successful learning algorithm. Our main results are two theorems that establish criteria for learnability for many classes of stochastic processes, including all special cases studied previously in the literature."}],"intvolume":" 54","title":"Learning theory for conditional risk minimization","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"1108","oa_version":"Submitted Version","article_processing_charge":"No","day":"01","page":"213 - 222","citation":{"chicago":"Zimin, Alexander, and Christoph Lampert. “Learning Theory for Conditional Risk Minimization,” 54:213–22. ML Research Press, 2017.","short":"A. Zimin, C. Lampert, in:, ML Research Press, 2017, pp. 213–222.","mla":"Zimin, Alexander, and Christoph Lampert. Learning Theory for Conditional Risk Minimization. Vol. 54, ML Research Press, 2017, pp. 213–22.","apa":"Zimin, A., & Lampert, C. (2017). Learning theory for conditional risk minimization (Vol. 54, pp. 213–222). Presented at the AISTATS: Artificial Intelligence and Statistics, Fort Lauderdale, FL, United States: ML Research Press.","ieee":"A. Zimin and C. Lampert, “Learning theory for conditional risk minimization,” presented at the AISTATS: Artificial Intelligence and Statistics, Fort Lauderdale, FL, United States, 2017, vol. 54, pp. 213–222.","ista":"Zimin A, Lampert C. 2017. Learning theory for conditional risk minimization. AISTATS: Artificial Intelligence and Statistics, PMLR, vol. 54, 213–222.","ama":"Zimin A, Lampert C. Learning theory for conditional risk minimization. In: Vol 54. ML Research Press; 2017:213-222."},"date_published":"2017-04-01T00:00:00Z","publist_id":"6261","ec_funded":1,"department":[{"_id":"ChLa"}],"publisher":"ML Research Press","publication_status":"published","year":"2017","volume":54,"date_updated":"2023-10-17T10:01:12Z","date_created":"2018-12-11T11:50:11Z","author":[{"first_name":"Alexander","last_name":"Zimin","id":"37099E9C-F248-11E8-B48F-1D18A9856A87","full_name":"Zimin, Alexander"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"month":"04","project":[{"grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425","name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7"}],"quality_controlled":"1","isi":1,"main_file_link":[{"open_access":"1","url":"http://proceedings.mlr.press/v54/zimin17a/zimin17a.pdf"}],"oa":1,"external_id":{"isi":["000509368500024"]},"language":[{"iso":"eng"}],"conference":{"end_date":"2017-04-22","start_date":"2017-04-20","location":"Fort Lauderdale, FL, United States","name":"AISTATS: Artificial Intelligence and Statistics"}},{"date_published":"2017-06-08T00:00:00Z","page":"2807 - 2816","citation":{"ama":"Pentina A, Lampert C. Multi-task learning with labeled and unlabeled tasks. In: Vol 70. ML Research Press; 2017:2807-2816.","apa":"Pentina, A., & Lampert, C. (2017). Multi-task learning with labeled and unlabeled tasks (Vol. 70, pp. 2807–2816). Presented at the ICML: International Conference on Machine Learning, Sydney, Australia: ML Research Press.","ieee":"A. Pentina and C. Lampert, “Multi-task learning with labeled and unlabeled tasks,” presented at the ICML: International Conference on Machine Learning, Sydney, Australia, 2017, vol. 70, pp. 2807–2816.","ista":"Pentina A, Lampert C. 2017. Multi-task learning with labeled and unlabeled tasks. ICML: International Conference on Machine Learning, PMLR, vol. 70, 2807–2816.","short":"A. Pentina, C. Lampert, in:, ML Research Press, 2017, pp. 2807–2816.","mla":"Pentina, Anastasia, and Christoph Lampert. Multi-Task Learning with Labeled and Unlabeled Tasks. Vol. 70, ML Research Press, 2017, pp. 2807–16.","chicago":"Pentina, Anastasia, and Christoph Lampert. “Multi-Task Learning with Labeled and Unlabeled Tasks,” 70:2807–16. ML Research Press, 2017."},"day":"08","article_processing_charge":"No","scopus_import":"1","oa_version":"Submitted Version","status":"public","title":"Multi-task learning with labeled and unlabeled tasks","intvolume":" 70","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"999","abstract":[{"lang":"eng","text":"In multi-task learning, a learner is given a collection of prediction tasks and needs to solve all of them. In contrast to previous work, which required that annotated training data must be available for all tasks, we consider a new setting, in which for some tasks, potentially most of them, only unlabeled training data is provided. Consequently, to solve all tasks, information must be transferred between tasks with labels and tasks without labels. Focusing on an instance-based transfer method we analyze two variants of this setting: when the set of labeled tasks is fixed, and when it can be actively selected by the learner. We state and prove a generalization bound that covers both scenarios and derive from it an algorithm for making the choice of labeled tasks (in the active case) and for transferring information between the tasks in a principled way. We also illustrate the effectiveness of the algorithm on synthetic and real data. "}],"alternative_title":["PMLR"],"type":"conference","language":[{"iso":"eng"}],"conference":{"start_date":"2017-08-06","location":"Sydney, Australia","end_date":"2017-08-11","name":"ICML: International Conference on Machine Learning"},"isi":1,"quality_controlled":"1","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"main_file_link":[{"url":"https://arxiv.org/abs/1602.06518","open_access":"1"}],"oa":1,"external_id":{"isi":["000683309502093"]},"month":"06","publication_identifier":{"isbn":["9781510855144"]},"date_created":"2018-12-11T11:49:37Z","date_updated":"2023-10-17T11:53:32Z","volume":70,"author":[{"id":"42E87FC6-F248-11E8-B48F-1D18A9856A87","last_name":"Pentina","first_name":"Anastasia","full_name":"Pentina, Anastasia"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"publication_status":"published","publisher":"ML Research Press","department":[{"_id":"ChLa"}],"year":"2017","publist_id":"6399","ec_funded":1},{"pubrep_id":"775","file":[{"access_level":"open_access","file_name":"IST-2017-775-v1+1_main.pdf","file_size":237111,"content_type":"application/pdf","creator":"system","relation":"main_file","file_id":"4961","date_updated":"2018-12-12T10:12:42Z","date_created":"2018-12-12T10:12:42Z"},{"file_name":"IST-2017-775-v1+2_supplementary.pdf","access_level":"open_access","creator":"system","content_type":"application/pdf","file_size":185818,"file_id":"4962","relation":"main_file","date_created":"2018-12-12T10:12:43Z","date_updated":"2018-12-12T10:12:43Z"}],"oa_version":"Published Version","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","_id":"1098","intvolume":" 29","title":"Lifelong learning with weighted majority votes","status":"public","ddc":["006"],"abstract":[{"lang":"eng","text":"Better understanding of the potential benefits of information transfer and representation learning is an important step towards the goal of building intelligent systems that are able to persist in the world and learn over time. In this work, we consider a setting where the learner encounters a stream of tasks but is able to retain only limited information from each encountered task, such as a learned predictor. In contrast to most previous works analyzing this scenario, we do not make any distributional assumptions on the task generating process. Instead, we formulate a complexity measure that captures the diversity of the observed tasks. We provide a lifelong learning algorithm with error guarantees for every observed task (rather than on average). We show sample complexity reductions in comparison to solving every task in isolation in terms of our task complexity measure. Further, our algorithmic framework can naturally be viewed as learning a representation from encountered tasks with a neural network."}],"type":"conference","alternative_title":["Advances in Neural Information Processing Systems"],"date_published":"2016-12-01T00:00:00Z","citation":{"short":"A. Pentina, R. Urner, in:, Neural Information Processing Systems, 2016, pp. 3619–3627.","mla":"Pentina, Anastasia, and Ruth Urner. Lifelong Learning with Weighted Majority Votes. Vol. 29, Neural Information Processing Systems, 2016, pp. 3619–27.","chicago":"Pentina, Anastasia, and Ruth Urner. “Lifelong Learning with Weighted Majority Votes,” 29:3619–27. Neural Information Processing Systems, 2016.","ama":"Pentina A, Urner R. Lifelong learning with weighted majority votes. In: Vol 29. Neural Information Processing Systems; 2016:3619-3627.","ieee":"A. Pentina and R. Urner, “Lifelong learning with weighted majority votes,” presented at the NIPS: Neural Information Processing Systems, Barcelona, Spain, 2016, vol. 29, pp. 3619–3627.","apa":"Pentina, A., & Urner, R. (2016). Lifelong learning with weighted majority votes (Vol. 29, pp. 3619–3627). Presented at the NIPS: Neural Information Processing Systems, Barcelona, Spain: Neural Information Processing Systems.","ista":"Pentina A, Urner R. 2016. Lifelong learning with weighted majority votes. NIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 29, 3619–3627."},"page":"3619-3627","has_accepted_license":"1","day":"01","scopus_import":1,"author":[{"first_name":"Anastasia","last_name":"Pentina","id":"42E87FC6-F248-11E8-B48F-1D18A9856A87","full_name":"Pentina, Anastasia"},{"full_name":"Urner, Ruth","first_name":"Ruth","last_name":"Urner"}],"volume":29,"date_updated":"2021-01-12T06:48:15Z","date_created":"2018-12-11T11:50:08Z","acknowledgement":"This work was in parts funded by the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no 308036.\r\n\r\n","year":"2016","department":[{"_id":"ChLa"}],"publisher":"Neural Information Processing Systems","publication_status":"published","ec_funded":1,"publist_id":"6277","file_date_updated":"2018-12-12T10:12:43Z","conference":{"location":"Barcelona, Spain","start_date":"2016-12-05","end_date":"2016-12-10","name":"NIPS: Neural Information Processing Systems"},"language":[{"iso":"eng"}],"oa":1,"project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"quality_controlled":"1","month":"12"},{"day":"01","scopus_import":1,"date_published":"2016-09-01T00:00:00Z","page":"92.1-92.12","citation":{"ama":"Kolesnikov A, Lampert C. Improving weakly-supervised object localization by micro-annotation. In: Proceedings of the British Machine Vision Conference 2016. Vol 2016-September. BMVA Press; 2016:92.1-92.12. doi:10.5244/C.30.92","ista":"Kolesnikov A, Lampert C. 2016. Improving weakly-supervised object localization by micro-annotation. Proceedings of the British Machine Vision Conference 2016. BMVC: British Machine Vision Conference vol. 2016–September, 92.1-92.12.","apa":"Kolesnikov, A., & Lampert, C. (2016). Improving weakly-supervised object localization by micro-annotation. In Proceedings of the British Machine Vision Conference 2016 (Vol. 2016–September, p. 92.1-92.12). York, United Kingdom: BMVA Press. https://doi.org/10.5244/C.30.92","ieee":"A. Kolesnikov and C. Lampert, “Improving weakly-supervised object localization by micro-annotation,” in Proceedings of the British Machine Vision Conference 2016, York, United Kingdom, 2016, vol. 2016–September, p. 92.1-92.12.","mla":"Kolesnikov, Alexander, and Christoph Lampert. “Improving Weakly-Supervised Object Localization by Micro-Annotation.” Proceedings of the British Machine Vision Conference 2016, vol. 2016–September, BMVA Press, 2016, p. 92.1-92.12, doi:10.5244/C.30.92.","short":"A. Kolesnikov, C. Lampert, in:, Proceedings of the British Machine Vision Conference 2016, BMVA Press, 2016, p. 92.1-92.12.","chicago":"Kolesnikov, Alexander, and Christoph Lampert. “Improving Weakly-Supervised Object Localization by Micro-Annotation.” In Proceedings of the British Machine Vision Conference 2016, 2016–September:92.1-92.12. BMVA Press, 2016. https://doi.org/10.5244/C.30.92."},"publication":"Proceedings of the British Machine Vision Conference 2016","abstract":[{"text":"Weakly-supervised object localization methods tend to fail for object classes that consistently co-occur with the same background elements, e.g. trains on tracks. We propose a method to overcome these failures by adding a very small amount of model-specific additional annotation. The main idea is to cluster a deep network\\'s mid-level representations and assign object or distractor labels to each cluster. Experiments show substantially improved localization results on the challenging ILSVC2014 dataset for bounding box detection and the PASCAL VOC2012 dataset for semantic segmentation.","lang":"eng"}],"type":"conference","oa_version":"Published Version","title":"Improving weakly-supervised object localization by micro-annotation","status":"public","_id":"1102","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","month":"09","language":[{"iso":"eng"}],"doi":"10.5244/C.30.92","conference":{"end_date":"2016-09-22","start_date":"2016-09-19","location":"York, United Kingdom","name":"BMVC: British Machine Vision Conference"},"project":[{"grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425","name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7"}],"quality_controlled":"1","main_file_link":[{"url":"http://www.bmva.org/bmvc/2016/papers/paper092/paper092.pdf","open_access":"1"}],"oa":1,"publist_id":"6273","ec_funded":1,"volume":"2016-September","date_created":"2018-12-11T11:50:09Z","date_updated":"2021-01-12T06:48:18Z","author":[{"first_name":"Alexander","last_name":"Kolesnikov","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","full_name":"Kolesnikov, Alexander"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"department":[{"_id":"ChLa"}],"publisher":"BMVA Press","publication_status":"published","year":"2016","acknowledgement":"This work was funded in parts by the European Research Council\r\nunder the European Union’s Seventh Framework Programme (FP7/2007-2013)/ERC grant\r\nagreement no 308036. We gratefully acknowledge the support of NVIDIA Corporation with\r\nthe donation of the GPUs used for this research."},{"type":"conference","article_number":"7759138","publist_id":"6121","abstract":[{"text":"With the accelerated development of robot technologies, optimal control becomes one of the central themes of research. In traditional approaches, the controller, by its internal functionality, finds appropriate actions on the basis of the history of sensor values, guided by the goals, intentions, objectives, learning schemes, and so forth. While very successful with classical robots, these methods run into severe difficulties when applied to soft robots, a new field of robotics with large interest for human-robot interaction. We claim that a novel controller paradigm opens new perspective for this field. This paper applies a recently developed neuro controller with differential extrinsic synaptic plasticity to a muscle-tendon driven arm-shoulder system from the Myorobotics toolkit. In the experiments, we observe a vast variety of self-organized behavior patterns: when left alone, the arm realizes pseudo-random sequences of different poses. By applying physical forces, the system can be entrained into definite motion patterns like wiping a table. Most interestingly, after attaching an object, the controller gets in a functional resonance with the object's internal dynamics, starting to shake spontaneously bottles half-filled with water or sensitively driving an attached pendulum into a circular mode. When attached to the crank of a wheel the neural system independently develops to rotate it. In this way, the robot discovers affordances of objects its body is interacting with.","lang":"eng"}],"department":[{"_id":"ChLa"},{"_id":"GaTk"}],"publisher":"IEEE","title":"Compliant control for soft robots: Emergent behavior of a tendon driven anthropomorphic arm","status":"public","publication_status":"published","_id":"1214","acknowledgement":"RD thanks for the hospitality at the Max-Planck-Institute and for helpful discussions with Nihat Ay and Keyan Zahedi.","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","year":"2016","oa_version":"None","volume":"2016-November","date_created":"2018-12-11T11:50:45Z","date_updated":"2021-01-12T06:49:08Z","author":[{"id":"3A276B68-F248-11E8-B48F-1D18A9856A87","first_name":"Georg S","last_name":"Martius","full_name":"Martius, Georg S"},{"first_name":"Raphael","last_name":"Hostettler","full_name":"Hostettler, Raphael"},{"full_name":"Knoll, Alois","first_name":"Alois","last_name":"Knoll"},{"first_name":"Ralf","last_name":"Der","full_name":"Der, Ralf"}],"scopus_import":1,"month":"11","day":"28","quality_controlled":"1","citation":{"ama":"Martius GS, Hostettler R, Knoll A, Der R. Compliant control for soft robots: Emergent behavior of a tendon driven anthropomorphic arm. In: Vol 2016-November. IEEE; 2016. doi:10.1109/IROS.2016.7759138","apa":"Martius, G. S., Hostettler, R., Knoll, A., & Der, R. (2016). Compliant control for soft robots: Emergent behavior of a tendon driven anthropomorphic arm (Vol. 2016–November). Presented at the IEEE RSJ International Conference on Intelligent Robots and Systems IROS , Daejeon, Korea: IEEE. https://doi.org/10.1109/IROS.2016.7759138","ieee":"G. S. Martius, R. Hostettler, A. Knoll, and R. Der, “Compliant control for soft robots: Emergent behavior of a tendon driven anthropomorphic arm,” presented at the IEEE RSJ International Conference on Intelligent Robots and Systems IROS , Daejeon, Korea, 2016, vol. 2016–November.","ista":"Martius GS, Hostettler R, Knoll A, Der R. 2016. Compliant control for soft robots: Emergent behavior of a tendon driven anthropomorphic arm. IEEE RSJ International Conference on Intelligent Robots and Systems IROS vol. 2016–November, 7759138.","short":"G.S. Martius, R. Hostettler, A. Knoll, R. Der, in:, IEEE, 2016.","mla":"Martius, Georg S., et al. Compliant Control for Soft Robots: Emergent Behavior of a Tendon Driven Anthropomorphic Arm. Vol. 2016–November, 7759138, IEEE, 2016, doi:10.1109/IROS.2016.7759138.","chicago":"Martius, Georg S, Raphael Hostettler, Alois Knoll, and Ralf Der. “Compliant Control for Soft Robots: Emergent Behavior of a Tendon Driven Anthropomorphic Arm,” Vol. 2016–November. IEEE, 2016. https://doi.org/10.1109/IROS.2016.7759138."},"language":[{"iso":"eng"}],"date_published":"2016-11-28T00:00:00Z","doi":"10.1109/IROS.2016.7759138","conference":{"name":"IEEE RSJ International Conference on Intelligent Robots and Systems IROS ","end_date":"2016-09-14","start_date":"2016-09-09","location":"Daejeon, Korea"}},{"abstract":[{"text":"We introduce a new loss function for the weakly-supervised training of semantic image segmentation models based on three guiding principles: to seed with weak localization cues, to expand objects based on the information about which classes can occur in an image, and to constrain the segmentations to coincide with object boundaries. We show experimentally that training a deep convolutional neural network using the proposed loss function leads to substantially better segmentations than previous state-of-the-art methods on the challenging PASCAL VOC 2012 dataset. We furthermore give insight into the working mechanism of our method by a detailed experimental study that illustrates how the segmentation quality is affected by each term of the proposed loss function as well as their combinations.","lang":"eng"}],"type":"conference","alternative_title":["LNCS"],"oa_version":"Preprint","_id":"1369","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","intvolume":" 9908","title":"Seed, expand and constrain: Three principles for weakly-supervised image segmentation","status":"public","day":"15","scopus_import":1,"date_published":"2016-09-15T00:00:00Z","citation":{"chicago":"Kolesnikov, Alexander, and Christoph Lampert. “Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation,” 9908:695–711. Springer, 2016. https://doi.org/10.1007/978-3-319-46493-0_42.","short":"A. Kolesnikov, C. Lampert, in:, Springer, 2016, pp. 695–711.","mla":"Kolesnikov, Alexander, and Christoph Lampert. Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation. Vol. 9908, Springer, 2016, pp. 695–711, doi:10.1007/978-3-319-46493-0_42.","apa":"Kolesnikov, A., & Lampert, C. (2016). Seed, expand and constrain: Three principles for weakly-supervised image segmentation (Vol. 9908, pp. 695–711). Presented at the ECCV: European Conference on Computer Vision, Amsterdam, The Netherlands: Springer. https://doi.org/10.1007/978-3-319-46493-0_42","ieee":"A. Kolesnikov and C. Lampert, “Seed, expand and constrain: Three principles for weakly-supervised image segmentation,” presented at the ECCV: European Conference on Computer Vision, Amsterdam, The Netherlands, 2016, vol. 9908, pp. 695–711.","ista":"Kolesnikov A, Lampert C. 2016. Seed, expand and constrain: Three principles for weakly-supervised image segmentation. ECCV: European Conference on Computer Vision, LNCS, vol. 9908, 695–711.","ama":"Kolesnikov A, Lampert C. Seed, expand and constrain: Three principles for weakly-supervised image segmentation. In: Vol 9908. Springer; 2016:695-711. doi:10.1007/978-3-319-46493-0_42"},"page":"695 - 711","ec_funded":1,"publist_id":"5842","author":[{"full_name":"Kolesnikov, Alexander","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","last_name":"Kolesnikov","first_name":"Alexander"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"volume":9908,"date_created":"2018-12-11T11:51:37Z","date_updated":"2021-01-12T06:50:12Z","year":"2016","publisher":"Springer","department":[{"_id":"ChLa"}],"publication_status":"published","month":"09","doi":"10.1007/978-3-319-46493-0_42","conference":{"end_date":"2016-10-14","location":"Amsterdam, The Netherlands","start_date":"2016-10-11","name":"ECCV: European Conference on Computer Vision"},"language":[{"iso":"eng"}],"main_file_link":[{"url":"https://arxiv.org/abs/1603.06098","open_access":"1"}],"oa":1,"project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"quality_controlled":"1"},{"author":[{"id":"49BC895A-F248-11E8-B48F-1D18A9856A87","last_name":"Pielorz","first_name":"Jasmin","full_name":"Pielorz, Jasmin"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"oa_version":"None","date_created":"2018-12-11T11:53:35Z","date_updated":"2021-01-12T06:52:39Z","_id":"1707","acknowledgement":"The DRIVER FP7 project has received funding from the European Unions Seventh Framework Programme for research, technological development and demonstration under grant agreement no 607798. RE-ACTA was funded within the framework of the Austrian Security Research Programme KIRAS by the Federal Ministry for Transport, Innovation and Technology.","year":"2016","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","department":[{"_id":"ChLa"}],"publisher":"IEEE","title":"Optimal geospatial allocation of volunteers for crisis management","publication_status":"published","status":"public","publist_id":"5429","abstract":[{"text":"Volunteer supporters play an important role in modern crisis and disaster management. In the times of mobile Internet devices, help from thousands of volunteers can be requested within a short time span, thus relieving professional helpers from minor chores or geographically spread-out tasks. However, the simultaneous availability of many volunteers also poses new problems. In particular, the volunteer efforts must be well coordinated, or otherwise situations might emerge in which too many idle volunteers at one location become more of a burden than a relief to the professionals.\r\nIn this work, we study the task of optimally assigning volunteers to selected locations, e.g. in order to perform regular measurements, to report on damage, or to distribute information or resources to the population in a crisis situation. We formulate the assignment tasks as an optimization problem and propose an effective and efficient solution procedure. Experiments on real data of the Team Österreich, consisting of over 36,000 Austrian volunteers, show the effectiveness and efficiency of our approach.","lang":"eng"}],"type":"conference","article_number":"7402041","doi":"10.1109/ICT-DM.2015.7402041","date_published":"2016-02-11T00:00:00Z","conference":{"name":"ICT-DM: Information and Communication Technologies for Disaster Management","end_date":"2015-12-02","location":"Rennes, France","start_date":"2015-11-30"},"language":[{"iso":"eng"}],"citation":{"chicago":"Pielorz, Jasmin, and Christoph Lampert. “Optimal Geospatial Allocation of Volunteers for Crisis Management.” IEEE, 2016. https://doi.org/10.1109/ICT-DM.2015.7402041.","short":"J. Pielorz, C. Lampert, in:, IEEE, 2016.","mla":"Pielorz, Jasmin, and Christoph Lampert. Optimal Geospatial Allocation of Volunteers for Crisis Management. 7402041, IEEE, 2016, doi:10.1109/ICT-DM.2015.7402041.","apa":"Pielorz, J., & Lampert, C. (2016). Optimal geospatial allocation of volunteers for crisis management. Presented at the ICT-DM: Information and Communication Technologies for Disaster Management, Rennes, France: IEEE. https://doi.org/10.1109/ICT-DM.2015.7402041","ieee":"J. Pielorz and C. Lampert, “Optimal geospatial allocation of volunteers for crisis management,” presented at the ICT-DM: Information and Communication Technologies for Disaster Management, Rennes, France, 2016.","ista":"Pielorz J, Lampert C. 2016. Optimal geospatial allocation of volunteers for crisis management. ICT-DM: Information and Communication Technologies for Disaster Management, 7402041.","ama":"Pielorz J, Lampert C. Optimal geospatial allocation of volunteers for crisis management. In: IEEE; 2016. doi:10.1109/ICT-DM.2015.7402041"},"quality_controlled":"1","month":"02","day":"11","scopus_import":1},{"publication_status":"published","publisher":"MIT Press","department":[{"_id":"ChLa"},{"_id":"GaTk"}],"year":"2016","date_updated":"2021-01-12T08:16:53Z","date_created":"2020-07-05T22:00:47Z","volume":28,"author":[{"full_name":"Martius, Georg S","id":"3A276B68-F248-11E8-B48F-1D18A9856A87","last_name":"Martius","first_name":"Georg S"},{"last_name":"Hostettler","first_name":"Rafael","full_name":"Hostettler, Rafael"},{"last_name":"Knoll","first_name":"Alois","full_name":"Knoll, Alois"},{"first_name":"Ralf","last_name":"Der","full_name":"Der, Ralf"}],"file_date_updated":"2020-07-14T12:48:09Z","ec_funded":1,"quality_controlled":"1","project":[{"call_identifier":"FP7","name":"International IST Postdoc Fellowship Programme","_id":"25681D80-B435-11E9-9278-68D0E5697425","grant_number":"291734"}],"tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"oa":1,"language":[{"iso":"eng"}],"conference":{"name":"ALIFE 2016: 15th International Conference on the Synthesis and Simulation of Living Systems","end_date":"2016-07-08","location":"Cancun, Mexico","start_date":"2016-07-04"},"doi":"10.7551/978-0-262-33936-0-ch029","month":"09","publication_identifier":{"isbn":["9780262339360"]},"status":"public","title":"Self-organized control of an tendon driven arm by differential extrinsic plasticity","ddc":["610"],"intvolume":" 28","_id":"8094","user_id":"D865714E-FA4E-11E9-B85B-F5C5E5697425","file":[{"date_updated":"2020-07-14T12:48:09Z","date_created":"2020-07-06T12:59:09Z","checksum":"cff63e7a4b8ac466ba51a9c84153a940","file_id":"8096","relation":"main_file","creator":"cziletti","file_size":678670,"content_type":"application/pdf","file_name":"2016_ProcALIFE_Martius.pdf","access_level":"open_access"}],"oa_version":"Published Version","type":"conference","abstract":[{"lang":"eng","text":"With the accelerated development of robot technologies, optimal control becomes one of the central themes of research. In traditional approaches, the controller, by its internal functionality, finds appropriate actions on the basis of the history of sensor values, guided by the goals, intentions, objectives, learning schemes, and so forth. The idea is that the controller controls the world---the body plus its environment---as reliably as possible. This paper focuses on new lines of self-organization for developmental robotics. We apply the recently developed differential extrinsic synaptic plasticity to a muscle-tendon driven arm-shoulder system from the Myorobotics toolkit. In the experiments, we observe a vast variety of self-organized behavior patterns: when left alone, the arm realizes pseudo-random sequences of different poses. By applying physical forces, the system can be entrained into definite motion patterns like wiping a table. Most interestingly, after attaching an object, the controller gets in a functional resonance with the object's internal dynamics, starting to shake spontaneously bottles half-filled with water or sensitively driving an attached pendulum into a circular mode. When attached to the crank of a wheel the neural system independently discovers how to rotate it. In this way, the robot discovers affordances of objects its body is interacting with."}],"page":"142-143","publication":"Proceedings of the Artificial Life Conference 2016","citation":{"chicago":"Martius, Georg S, Rafael Hostettler, Alois Knoll, and Ralf Der. “Self-Organized Control of an Tendon Driven Arm by Differential Extrinsic Plasticity.” In Proceedings of the Artificial Life Conference 2016, 28:142–43. MIT Press, 2016. https://doi.org/10.7551/978-0-262-33936-0-ch029.","mla":"Martius, Georg S., et al. “Self-Organized Control of an Tendon Driven Arm by Differential Extrinsic Plasticity.” Proceedings of the Artificial Life Conference 2016, vol. 28, MIT Press, 2016, pp. 142–43, doi:10.7551/978-0-262-33936-0-ch029.","short":"G.S. Martius, R. Hostettler, A. Knoll, R. Der, in:, Proceedings of the Artificial Life Conference 2016, MIT Press, 2016, pp. 142–143.","ista":"Martius GS, Hostettler R, Knoll A, Der R. 2016. Self-organized control of an tendon driven arm by differential extrinsic plasticity. Proceedings of the Artificial Life Conference 2016. ALIFE 2016: 15th International Conference on the Synthesis and Simulation of Living Systems vol. 28, 142–143.","ieee":"G. S. Martius, R. Hostettler, A. Knoll, and R. Der, “Self-organized control of an tendon driven arm by differential extrinsic plasticity,” in Proceedings of the Artificial Life Conference 2016, Cancun, Mexico, 2016, vol. 28, pp. 142–143.","apa":"Martius, G. S., Hostettler, R., Knoll, A., & Der, R. (2016). Self-organized control of an tendon driven arm by differential extrinsic plasticity. In Proceedings of the Artificial Life Conference 2016 (Vol. 28, pp. 142–143). Cancun, Mexico: MIT Press. https://doi.org/10.7551/978-0-262-33936-0-ch029","ama":"Martius GS, Hostettler R, Knoll A, Der R. Self-organized control of an tendon driven arm by differential extrinsic plasticity. In: Proceedings of the Artificial Life Conference 2016. Vol 28. MIT Press; 2016:142-143. doi:10.7551/978-0-262-33936-0-ch029"},"date_published":"2016-09-01T00:00:00Z","scopus_import":1,"day":"01","has_accepted_license":"1","article_processing_charge":"No"},{"publication_identifier":{"issn":["2663-337X"]},"month":"11","language":[{"iso":"eng"}],"degree_awarded":"PhD","supervisor":[{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"doi":"10.15479/AT:ISTA:TH_776","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"oa":1,"publist_id":"6234","ec_funded":1,"file_date_updated":"2018-12-12T10:14:07Z","date_created":"2018-12-11T11:50:17Z","date_updated":"2023-09-07T11:52:03Z","author":[{"id":"42E87FC6-F248-11E8-B48F-1D18A9856A87","last_name":"Pentina","first_name":"Anastasia","full_name":"Pentina, Anastasia"}],"department":[{"_id":"ChLa"}],"publisher":"Institute of Science and Technology Austria","publication_status":"published","year":"2016","acknowledgement":"First and foremost I would like to express my gratitude to my supervisor, Christoph\r\nLampert. Thank you for your patience in teaching me all aspects of doing research\r\n(including English grammar), for your trust in my capabilities and endless support. Thank\r\nyou for granting me freedom in my research and, at the same time, having time and\r\nhelping me cope with the consequences whenever I needed it. Thank you for creating\r\nan excellent atmosphere in the group, it was a great pleasure and honor to be a part of\r\nit. There could not have been a better and more inspiring adviser and mentor.\r\nI thank Shai Ben-David for welcoming me into his group at the University of Waterloo,\r\nfor inspiring discussions and support. It was a great pleasure to work together. I am\r\nalso thankful to Ruth Urner for hosting me at the Max-Planck Institute Tübingen, for the\r\nfruitful collaboration and for taking care of me during that not-so-sunny month of May.\r\nI thank Jan Maas for kindly joining my thesis committee despite the short notice and\r\nproviding me with insightful comments.\r\nI would like to thank my colleagues for their support, entertaining conversations and\r\nendless table soccer games we shared together: Georg, Jan, Amelie and Emilie, Michal\r\nand Alex, Alex K. and Alex Z., Thomas, Sameh, Vlad, Mayu, Nathaniel, Silvester, Neel,\r\nCsaba, Vladimir, Morten. Thank you, Mabel and Ram, for the wonderful time we spent\r\ntogether. I am thankful to Shrinu and Samira for taking care of me during my stay at the\r\nUniversity of Waterloo. Special thanks to Viktoriia for her never-ending optimism and for\r\nbeing so inspiring and supportive, especially at the beginning of my PhD journey.\r\nThanks to IST administration, in particular, Vlad and Elisabeth for shielding me from\r\nmost of the bureaucratic paperwork.\r\n\r\nThis dissertation would not have been possible without funding from the European\r\nResearch Council under the European Union's Seventh Framework Programme\r\n(FP7/2007-2013)/ERC grant agreement no 308036.","has_accepted_license":"1","article_processing_charge":"No","day":"01","date_published":"2016-11-01T00:00:00Z","page":"127","citation":{"short":"A. Pentina, Theoretical Foundations of Multi-Task Lifelong Learning, Institute of Science and Technology Austria, 2016.","mla":"Pentina, Anastasia. Theoretical Foundations of Multi-Task Lifelong Learning. Institute of Science and Technology Austria, 2016, doi:10.15479/AT:ISTA:TH_776.","chicago":"Pentina, Anastasia. “Theoretical Foundations of Multi-Task Lifelong Learning.” Institute of Science and Technology Austria, 2016. https://doi.org/10.15479/AT:ISTA:TH_776.","ama":"Pentina A. Theoretical foundations of multi-task lifelong learning. 2016. doi:10.15479/AT:ISTA:TH_776","ieee":"A. Pentina, “Theoretical foundations of multi-task lifelong learning,” Institute of Science and Technology Austria, 2016.","apa":"Pentina, A. (2016). Theoretical foundations of multi-task lifelong learning. Institute of Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:TH_776","ista":"Pentina A. 2016. Theoretical foundations of multi-task lifelong learning. Institute of Science and Technology Austria."},"abstract":[{"text":"Traditionally machine learning has been focusing on the problem of solving a single\r\ntask in isolation. While being quite well understood, this approach disregards an\r\nimportant aspect of human learning: when facing a new problem, humans are able to\r\nexploit knowledge acquired from previously learned tasks. Intuitively, access to several\r\nproblems simultaneously or sequentially could also be advantageous for a machine\r\nlearning system, especially if these tasks are closely related. Indeed, results of many\r\nempirical studies have provided justification for this intuition. However, theoretical\r\njustifications of this idea are rather limited.\r\nThe focus of this thesis is to expand the understanding of potential benefits of information\r\ntransfer between several related learning problems. We provide theoretical\r\nanalysis for three scenarios of multi-task learning - multiple kernel learning, sequential\r\nlearning and active task selection. We also provide a PAC-Bayesian perspective on\r\nlifelong learning and investigate how the task generation process influences the generalization\r\nguarantees in this scenario. In addition, we show how some of the obtained\r\ntheoretical results can be used to derive principled multi-task and lifelong learning\r\nalgorithms and illustrate their performance on various synthetic and real-world datasets.","lang":"eng"}],"alternative_title":["ISTA Thesis"],"type":"dissertation","file":[{"access_level":"open_access","file_name":"IST-2017-776-v1+1_Pentina_Thesis_2016.pdf","creator":"system","content_type":"application/pdf","file_size":2140062,"file_id":"5056","relation":"main_file","date_created":"2018-12-12T10:14:07Z","date_updated":"2018-12-12T10:14:07Z"}],"oa_version":"Published Version","pubrep_id":"776","ddc":["006"],"title":"Theoretical foundations of multi-task lifelong learning","status":"public","_id":"1126","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1"},{"month":"01","language":[{"iso":"eng"}],"conference":{"name":"NIPS: Neural Information Processing Systems","start_date":"2015-12-07","location":"Montreal, Canada","end_date":"2015-12-12"},"project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"quality_controlled":"1","main_file_link":[{"url":"http://papers.nips.cc/paper/6007-lifelong-learning-with-non-iid-tasks","open_access":"1"}],"oa":1,"publist_id":"5781","ec_funded":1,"volume":2015,"date_updated":"2021-01-12T06:50:39Z","date_created":"2018-12-11T11:51:57Z","author":[{"full_name":"Pentina, Anastasia","id":"42E87FC6-F248-11E8-B48F-1D18A9856A87","first_name":"Anastasia","last_name":"Pentina"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"department":[{"_id":"ChLa"}],"publisher":"Neural Information Processing Systems","publication_status":"published","year":"2015","day":"01","scopus_import":1,"date_published":"2015-01-01T00:00:00Z","page":"1540 - 1548","citation":{"short":"A. Pentina, C. Lampert, in:, Neural Information Processing Systems, 2015, pp. 1540–1548.","mla":"Pentina, Anastasia, and Christoph Lampert. Lifelong Learning with Non-i.i.d. Tasks. Vol. 2015, Neural Information Processing Systems, 2015, pp. 1540–48.","chicago":"Pentina, Anastasia, and Christoph Lampert. “Lifelong Learning with Non-i.i.d. Tasks,” 2015:1540–48. Neural Information Processing Systems, 2015.","ama":"Pentina A, Lampert C. Lifelong learning with non-i.i.d. tasks. In: Vol 2015. Neural Information Processing Systems; 2015:1540-1548.","ieee":"A. Pentina and C. Lampert, “Lifelong learning with non-i.i.d. tasks,” presented at the NIPS: Neural Information Processing Systems, Montreal, Canada, 2015, vol. 2015, pp. 1540–1548.","apa":"Pentina, A., & Lampert, C. (2015). Lifelong learning with non-i.i.d. tasks (Vol. 2015, pp. 1540–1548). Presented at the NIPS: Neural Information Processing Systems, Montreal, Canada: Neural Information Processing Systems.","ista":"Pentina A, Lampert C. 2015. Lifelong learning with non-i.i.d. tasks. NIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 2015, 1540–1548."},"abstract":[{"text":"In this work we aim at extending the theoretical foundations of lifelong learning. Previous work analyzing this scenario is based on the assumption that learning tasks are sampled i.i.d. from a task environment or limited to strongly constrained data distributions. Instead, we study two scenarios when lifelong learning is possible, even though the observed tasks do not form an i.i.d. sample: first, when they are sampled from the same environment, but possibly with dependencies, and second, when the task environment is allowed to change over time in a consistent way. In the first case we prove a PAC-Bayesian theorem that can be seen as a direct generalization of the analogous previous result for the i.i.d. case. For the second scenario we propose to learn an inductive bias in form of a transfer procedure. We present a generalization bound and show on a toy example how it can be used to identify a beneficial transfer algorithm.","lang":"eng"}],"alternative_title":["Advances in Neural Information Processing Systems"],"type":"conference","oa_version":"None","intvolume":" 2015","status":"public","title":"Lifelong learning with non-i.i.d. tasks","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","_id":"1425"},{"language":[{"iso":"eng"}],"doi":"10.1109/TCSVT.2014.2379972","date_published":"2015-08-01T00:00:00Z","quality_controlled":"1","page":"1295 - 1308","publication":"IEEE Transactions on Circuits and Systems for Video Technology","citation":{"short":"W. Xia, C. Domokos, J. Xiong, L. Cheong, S. Yan, IEEE Transactions on Circuits and Systems for Video Technology 25 (2015) 1295–1308.","mla":"Xia, Wei, et al. “Segmentation over Detection via Optimal Sparse Reconstructions.” IEEE Transactions on Circuits and Systems for Video Technology, vol. 25, no. 8, IEEE, 2015, pp. 1295–308, doi:10.1109/TCSVT.2014.2379972.","chicago":"Xia, Wei, Csaba Domokos, Junjun Xiong, Loongfah Cheong, and Shuicheng Yan. “Segmentation over Detection via Optimal Sparse Reconstructions.” IEEE Transactions on Circuits and Systems for Video Technology. IEEE, 2015. https://doi.org/10.1109/TCSVT.2014.2379972.","ama":"Xia W, Domokos C, Xiong J, Cheong L, Yan S. Segmentation over detection via optimal sparse reconstructions. IEEE Transactions on Circuits and Systems for Video Technology. 2015;25(8):1295-1308. doi:10.1109/TCSVT.2014.2379972","ieee":"W. Xia, C. Domokos, J. Xiong, L. Cheong, and S. Yan, “Segmentation over detection via optimal sparse reconstructions,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 25, no. 8. IEEE, pp. 1295–1308, 2015.","apa":"Xia, W., Domokos, C., Xiong, J., Cheong, L., & Yan, S. (2015). Segmentation over detection via optimal sparse reconstructions. IEEE Transactions on Circuits and Systems for Video Technology. IEEE. https://doi.org/10.1109/TCSVT.2014.2379972","ista":"Xia W, Domokos C, Xiong J, Cheong L, Yan S. 2015. Segmentation over detection via optimal sparse reconstructions. IEEE Transactions on Circuits and Systems for Video Technology. 25(8), 1295–1308."},"month":"08","day":"01","scopus_import":1,"date_created":"2018-12-11T11:52:34Z","date_updated":"2021-01-12T06:51:26Z","volume":25,"oa_version":"None","author":[{"last_name":"Xia","first_name":"Wei","full_name":"Xia, Wei"},{"id":"492DACF8-F248-11E8-B48F-1D18A9856A87","first_name":"Csaba","last_name":"Domokos","full_name":"Domokos, Csaba"},{"full_name":"Xiong, Junjun","first_name":"Junjun","last_name":"Xiong"},{"last_name":"Cheong","first_name":"Loongfah","full_name":"Cheong, Loongfah"},{"full_name":"Yan, Shuicheng","first_name":"Shuicheng","last_name":"Yan"}],"publication_status":"published","title":"Segmentation over detection via optimal sparse reconstructions","status":"public","intvolume":" 25","department":[{"_id":"ChLa"}],"publisher":"IEEE","year":"2015","_id":"1533","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"lang":"eng","text":"This paper addresses the problem of semantic segmentation, where the possible class labels are from a predefined set. We exploit top-down guidance, i.e., the coarse localization of the objects and their class labels provided by object detectors. For each detected bounding box, figure-ground segmentation is performed and the final result is achieved by merging the figure-ground segmentations. The main idea of the proposed approach, which is presented in our preliminary work, is to reformulate the figure-ground segmentation problem as sparse reconstruction pursuing the object mask in a nonparametric manner. The latent segmentation mask should be coherent subject to sparse error caused by intra-category diversity; thus, the object mask is inferred by making use of sparse representations over the training set. To handle local spatial deformations, local patch-level masks are also considered and inferred by sparse representations over the spatially nearby patches. The sparse reconstruction coefficients and the latent mask are alternately optimized by applying the Lasso algorithm and the accelerated proximal gradient method. The proposed formulation results in a convex optimization problem; thus, the global optimal solution is achieved. In this paper, we provide theoretical analysis of the convergence and optimality. We also give an extended numerical analysis of the proposed algorithm and a comprehensive comparison with the related semantic segmentation methods on the challenging PASCAL visual object class object segmentation datasets and the Weizmann horse dataset. The experimental results demonstrate that the proposed algorithm achieves a competitive performance when compared with the state of the arts."}],"publist_id":"5638","issue":"8","type":"journal_article"},{"type":"journal_article","issue":"45","abstract":[{"text":"Grounding autonomous behavior in the nervous system is a fundamental challenge for neuroscience. In particular, self-organized behavioral development provides more questions than answers. Are there special functional units for curiosity, motivation, and creativity? This paper argues that these features can be grounded in synaptic plasticity itself, without requiring any higher-level constructs. We propose differential extrinsic plasticity (DEP) as a new synaptic rule for self-learning systems and apply it to a number of complex robotic systems as a test case. Without specifying any purpose or goal, seemingly purposeful and adaptive rhythmic behavior is developed, displaying a certain level of sensorimotor intelligence. These surprising results require no systemspecific modifications of the DEP rule. They rather arise from the underlying mechanism of spontaneous symmetry breaking,which is due to the tight brain body environment coupling. The new synaptic rule is biologically plausible and would be an interesting target for neurobiological investigation. We also argue that this neuronal mechanism may have been a catalyst in natural evolution.","lang":"eng"}],"intvolume":" 112","title":"Novel plasticity rule can explain the development of sensorimotor intelligence","status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"1570","oa_version":"Submitted Version","scopus_import":1,"day":"10","page":"E6224 - E6232","citation":{"short":"R. Der, G.S. Martius, PNAS 112 (2015) E6224–E6232.","mla":"Der, Ralf, and Georg S. Martius. “Novel Plasticity Rule Can Explain the Development of Sensorimotor Intelligence.” PNAS, vol. 112, no. 45, National Academy of Sciences, 2015, pp. E6224–32, doi:10.1073/pnas.1508400112.","chicago":"Der, Ralf, and Georg S Martius. “Novel Plasticity Rule Can Explain the Development of Sensorimotor Intelligence.” PNAS. National Academy of Sciences, 2015. https://doi.org/10.1073/pnas.1508400112.","ama":"Der R, Martius GS. Novel plasticity rule can explain the development of sensorimotor intelligence. PNAS. 2015;112(45):E6224-E6232. doi:10.1073/pnas.1508400112","apa":"Der, R., & Martius, G. S. (2015). Novel plasticity rule can explain the development of sensorimotor intelligence. PNAS. National Academy of Sciences. https://doi.org/10.1073/pnas.1508400112","ieee":"R. Der and G. S. Martius, “Novel plasticity rule can explain the development of sensorimotor intelligence,” PNAS, vol. 112, no. 45. National Academy of Sciences, pp. E6224–E6232, 2015.","ista":"Der R, Martius GS. 2015. Novel plasticity rule can explain the development of sensorimotor intelligence. PNAS. 112(45), E6224–E6232."},"publication":"PNAS","date_published":"2015-11-10T00:00:00Z","publist_id":"5601","ec_funded":1,"department":[{"_id":"ChLa"},{"_id":"GaTk"}],"publisher":"National Academy of Sciences","publication_status":"published","pmid":1,"year":"2015","volume":112,"date_updated":"2021-01-12T06:51:40Z","date_created":"2018-12-11T11:52:47Z","author":[{"first_name":"Ralf","last_name":"Der","full_name":"Der, Ralf"},{"full_name":"Martius, Georg S","id":"3A276B68-F248-11E8-B48F-1D18A9856A87","first_name":"Georg S","last_name":"Martius"}],"month":"11","project":[{"grant_number":"291734","_id":"25681D80-B435-11E9-9278-68D0E5697425","call_identifier":"FP7","name":"International IST Postdoc Fellowship Programme"}],"quality_controlled":"1","main_file_link":[{"url":"http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4653169/","open_access":"1"}],"oa":1,"external_id":{"pmid":["26504200"]},"language":[{"iso":"eng"}],"doi":"10.1073/pnas.1508400112"},{"project":[{"_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036","call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding"}],"quality_controlled":"1","main_file_link":[{"url":"http://arxiv.org/abs/1602.06531","open_access":"1"}],"oa":1,"language":[{"iso":"eng"}],"doi":"10.1007/978-3-319-24486-0_13","conference":{"name":"ALT: Algorithmic Learning Theory","start_date":"2015-10-04","location":"Banff, AB, Canada","end_date":"2015-10-06"},"month":"01","department":[{"_id":"ChLa"}],"publisher":"Springer","publication_status":"published","year":"2015","volume":9355,"date_created":"2018-12-11T11:53:35Z","date_updated":"2021-01-12T06:52:39Z","author":[{"last_name":"Pentina","first_name":"Anastasia","id":"42E87FC6-F248-11E8-B48F-1D18A9856A87","full_name":"Pentina, Anastasia"},{"first_name":"Shai","last_name":"Ben David","full_name":"Ben David, Shai"}],"ec_funded":1,"publist_id":"5430","page":"194 - 208","citation":{"ama":"Pentina A, Ben David S. Multi-task and lifelong learning of kernels. In: Vol 9355. Springer; 2015:194-208. doi:10.1007/978-3-319-24486-0_13","ista":"Pentina A, Ben David S. 2015. Multi-task and lifelong learning of kernels. ALT: Algorithmic Learning Theory, LNCS, vol. 9355, 194–208.","apa":"Pentina, A., & Ben David, S. (2015). Multi-task and lifelong learning of kernels (Vol. 9355, pp. 194–208). Presented at the ALT: Algorithmic Learning Theory, Banff, AB, Canada: Springer. https://doi.org/10.1007/978-3-319-24486-0_13","ieee":"A. Pentina and S. Ben David, “Multi-task and lifelong learning of kernels,” presented at the ALT: Algorithmic Learning Theory, Banff, AB, Canada, 2015, vol. 9355, pp. 194–208.","mla":"Pentina, Anastasia, and Shai Ben David. Multi-Task and Lifelong Learning of Kernels. Vol. 9355, Springer, 2015, pp. 194–208, doi:10.1007/978-3-319-24486-0_13.","short":"A. Pentina, S. Ben David, in:, Springer, 2015, pp. 194–208.","chicago":"Pentina, Anastasia, and Shai Ben David. “Multi-Task and Lifelong Learning of Kernels,” 9355:194–208. Springer, 2015. https://doi.org/10.1007/978-3-319-24486-0_13."},"date_published":"2015-01-01T00:00:00Z","scopus_import":1,"day":"01","intvolume":" 9355","status":"public","title":"Multi-task and lifelong learning of kernels","_id":"1706","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","oa_version":"Preprint","alternative_title":["LNCS"],"type":"conference","abstract":[{"lang":"eng","text":"We consider a problem of learning kernels for use in SVM classification in the multi-task and lifelong scenarios and provide generalization bounds on the error of a large margin classifier. Our results show that, under mild conditions on the family of kernels used for learning, solving several related tasks simultaneously is beneficial over single task learning. In particular, as the number of observed tasks grows, assuming that in the considered family of kernels there exists one that yields low approximation error on all tasks, the overhead associated with learning such a kernel vanishes and the complexity converges to that of learning when this good kernel is given to the learner."}]},{"day":"01","month":"06","scopus_import":1,"conference":{"start_date":"2015-06-07","location":"Boston, MA, USA","end_date":"2015-06-12","name":"CVPR: Computer Vision and Pattern Recognition"},"doi":"10.1109/CVPR.2015.7298890","date_published":"2015-06-01T00:00:00Z","language":[{"iso":"eng"}],"main_file_link":[{"open_access":"1","url":"http://arxiv.org/abs/1408.6804"}],"oa":1,"citation":{"chicago":"Shah, Neel, Vladimir Kolmogorov, and Christoph Lampert. “A Multi-Plane Block-Coordinate Frank-Wolfe Algorithm for Training Structural SVMs with a Costly Max-Oracle,” 2737–45. IEEE, 2015. https://doi.org/10.1109/CVPR.2015.7298890.","mla":"Shah, Neel, et al. A Multi-Plane Block-Coordinate Frank-Wolfe Algorithm for Training Structural SVMs with a Costly Max-Oracle. IEEE, 2015, pp. 2737–45, doi:10.1109/CVPR.2015.7298890.","short":"N. Shah, V. Kolmogorov, C. Lampert, in:, IEEE, 2015, pp. 2737–2745.","ista":"Shah N, Kolmogorov V, Lampert C. 2015. A multi-plane block-coordinate Frank-Wolfe algorithm for training structural SVMs with a costly max-oracle. CVPR: Computer Vision and Pattern Recognition, 2737–2745.","ieee":"N. Shah, V. Kolmogorov, and C. Lampert, “A multi-plane block-coordinate Frank-Wolfe algorithm for training structural SVMs with a costly max-oracle,” presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, USA, 2015, pp. 2737–2745.","apa":"Shah, N., Kolmogorov, V., & Lampert, C. (2015). A multi-plane block-coordinate Frank-Wolfe algorithm for training structural SVMs with a costly max-oracle (pp. 2737–2745). Presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, USA: IEEE. https://doi.org/10.1109/CVPR.2015.7298890","ama":"Shah N, Kolmogorov V, Lampert C. A multi-plane block-coordinate Frank-Wolfe algorithm for training structural SVMs with a costly max-oracle. In: IEEE; 2015:2737-2745. doi:10.1109/CVPR.2015.7298890"},"quality_controlled":"1","page":"2737 - 2745","project":[{"_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036","call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding"},{"name":"Discrete Optimization in Computer Vision: Theory and Practice","call_identifier":"FP7","_id":"25FBA906-B435-11E9-9278-68D0E5697425","grant_number":"616160"}],"abstract":[{"lang":"eng","text":"Structural support vector machines (SSVMs) are amongst the best performing models for structured computer vision tasks, such as semantic image segmentation or human pose estimation. Training SSVMs, however, is computationally costly, because it requires repeated calls to a structured prediction subroutine (called \\emph{max-oracle}), which has to solve an optimization problem itself, e.g. a graph cut.\r\nIn this work, we introduce a new algorithm for SSVM training that is more efficient than earlier techniques when the max-oracle is computationally expensive, as it is frequently the case in computer vision tasks. The main idea is to (i) combine the recent stochastic Block-Coordinate Frank-Wolfe algorithm with efficient hyperplane caching, and (ii) use an automatic selection rule for deciding whether to call the exact max-oracle or to rely on an approximate one based on the cached hyperplanes.\r\nWe show experimentally that this strategy leads to faster convergence to the optimum with respect to the number of requires oracle calls, and that this translates into faster convergence with respect to the total runtime when the max-oracle is slow compared to the other steps of the algorithm. "}],"publist_id":"5240","ec_funded":1,"type":"conference","author":[{"id":"31ABAF80-F248-11E8-B48F-1D18A9856A87","first_name":"Neel","last_name":"Shah","full_name":"Shah, Neel"},{"full_name":"Kolmogorov, Vladimir","last_name":"Kolmogorov","first_name":"Vladimir","id":"3D50B0BA-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"date_updated":"2021-01-12T06:53:40Z","date_created":"2018-12-11T11:54:24Z","oa_version":"Preprint","year":"2015","_id":"1859","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","title":"A multi-plane block-coordinate Frank-Wolfe algorithm for training structural SVMs with a costly max-oracle","status":"public","publication_status":"published","publisher":"IEEE","department":[{"_id":"VlKo"},{"_id":"ChLa"}]},{"scopus_import":1,"day":"01","month":"06","main_file_link":[{"url":"http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Royer_Classifier_Adaptation_at_2015_CVPR_paper.pdf","open_access":"1"}],"citation":{"chicago":"Royer, Amélie, and Christoph Lampert. “Classifier Adaptation at Prediction Time,” 1401–9. IEEE, 2015. https://doi.org/10.1109/CVPR.2015.7298746.","mla":"Royer, Amélie, and Christoph Lampert. Classifier Adaptation at Prediction Time. IEEE, 2015, pp. 1401–09, doi:10.1109/CVPR.2015.7298746.","short":"A. Royer, C. Lampert, in:, IEEE, 2015, pp. 1401–1409.","ista":"Royer A, Lampert C. 2015. Classifier adaptation at prediction time. CVPR: Computer Vision and Pattern Recognition, 1401–1409.","apa":"Royer, A., & Lampert, C. (2015). Classifier adaptation at prediction time (pp. 1401–1409). Presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, United States: IEEE. https://doi.org/10.1109/CVPR.2015.7298746","ieee":"A. Royer and C. Lampert, “Classifier adaptation at prediction time,” presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, United States, 2015, pp. 1401–1409.","ama":"Royer A, Lampert C. Classifier adaptation at prediction time. In: IEEE; 2015:1401-1409. doi:10.1109/CVPR.2015.7298746"},"oa":1,"quality_controlled":"1","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"page":"1401 - 1409","conference":{"name":"CVPR: Computer Vision and Pattern Recognition","end_date":"2015-06-12","start_date":"2015-06-07","location":"Boston, MA, United States"},"date_published":"2015-06-01T00:00:00Z","doi":"10.1109/CVPR.2015.7298746","language":[{"iso":"eng"}],"type":"conference","abstract":[{"lang":"eng","text":"Classifiers for object categorization are usually evaluated by their accuracy on a set of i.i.d. test examples. This provides us with an estimate of the expected error when applying the classifiers to a single new image. In real application, however, classifiers are rarely only used for a single image and then discarded. Instead, they are applied sequentially to many images, and these are typically not i.i.d. samples from a fixed data distribution, but they carry dependencies and their class distribution varies over time. In this work, we argue that the phenomenon of correlated data at prediction time is not a nuisance, but a blessing in disguise. We describe a probabilistic method for adapting classifiers at prediction time without having to retrain them. We also introduce a framework for creating realistically distributed image sequences, which offers a way to benchmark classifier adaptation methods, such as the one we propose. Experiments on the ILSVRC2010 and ILSVRC2012 datasets show that adapting object classification systems at prediction time can significantly reduce their error rate, even with no additional human feedback."}],"ec_funded":1,"publist_id":"5239","year":"2015","_id":"1860","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","status":"public","publication_status":"published","title":"Classifier adaptation at prediction time","department":[{"_id":"ChLa"}],"publisher":"IEEE","author":[{"full_name":"Royer, Amélie","last_name":"Royer","first_name":"Amélie"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2018-12-11T11:54:24Z","date_updated":"2021-01-12T06:53:41Z","oa_version":"Submitted Version"},{"scopus_import":1,"day":"15","month":"10","oa":1,"external_id":{"arxiv":["1406.5362"]},"citation":{"mla":"Lampert, Christoph. Predicting the Future Behavior of a Time-Varying Probability Distribution. IEEE, 2015, pp. 942–50, doi:10.1109/CVPR.2015.7298696.","short":"C. Lampert, in:, IEEE, 2015, pp. 942–950.","chicago":"Lampert, Christoph. “Predicting the Future Behavior of a Time-Varying Probability Distribution,” 942–50. IEEE, 2015. https://doi.org/10.1109/CVPR.2015.7298696.","ama":"Lampert C. Predicting the future behavior of a time-varying probability distribution. In: IEEE; 2015:942-950. doi:10.1109/CVPR.2015.7298696","ista":"Lampert C. 2015. Predicting the future behavior of a time-varying probability distribution. CVPR: Computer Vision and Pattern Recognition, 942–950.","ieee":"C. Lampert, “Predicting the future behavior of a time-varying probability distribution,” presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, United States, 2015, pp. 942–950.","apa":"Lampert, C. (2015). Predicting the future behavior of a time-varying probability distribution (pp. 942–950). Presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, United States: IEEE. https://doi.org/10.1109/CVPR.2015.7298696"},"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1406.5362"}],"quality_controlled":"1","page":"942 - 950","conference":{"name":"CVPR: Computer Vision and Pattern Recognition","end_date":"2015-06-12","location":"Boston, MA, United States","start_date":"2015-06-07"},"doi":"10.1109/CVPR.2015.7298696","date_published":"2015-10-15T00:00:00Z","language":[{"iso":"eng"}],"type":"conference","abstract":[{"text":"We study the problem of predicting the future, though only in the probabilistic sense of estimating a future state of a time-varying probability distribution. This is not only an interesting academic problem, but solving this extrapolation problem also has many practical application, e.g. for training classifiers that have to operate under time-varying conditions. Our main contribution is a method for predicting the next step of the time-varying distribution from a given sequence of sample sets from earlier time steps. For this we rely on two recent machine learning techniques: embedding probability distributions into a reproducing kernel Hilbert space, and learning operators by vector-valued regression. We illustrate the working principles and the practical usefulness of our method by experiments on synthetic and real data. We also highlight an exemplary application: training a classifier in a domain adaptation setting without having access to examples from the test time distribution at training time.","lang":"eng"}],"publist_id":"5241","year":"2015","_id":"1858","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","title":"Predicting the future behavior of a time-varying probability distribution","status":"public","publication_status":"published","publisher":"IEEE","department":[{"_id":"ChLa"}],"author":[{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2018-12-11T11:54:24Z","date_updated":"2021-01-12T06:53:40Z","oa_version":"Preprint"},{"oa":1,"citation":{"ama":"Pentina A, Sharmanska V, Lampert C. Curriculum learning of multiple tasks. In: IEEE; 2015:5492-5500. doi:10.1109/CVPR.2015.7299188","apa":"Pentina, A., Sharmanska, V., & Lampert, C. (2015). Curriculum learning of multiple tasks (pp. 5492–5500). Presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, United States: IEEE. https://doi.org/10.1109/CVPR.2015.7299188","ieee":"A. Pentina, V. Sharmanska, and C. Lampert, “Curriculum learning of multiple tasks,” presented at the CVPR: Computer Vision and Pattern Recognition, Boston, MA, United States, 2015, pp. 5492–5500.","ista":"Pentina A, Sharmanska V, Lampert C. 2015. Curriculum learning of multiple tasks. CVPR: Computer Vision and Pattern Recognition, 5492–5500.","short":"A. Pentina, V. Sharmanska, C. Lampert, in:, IEEE, 2015, pp. 5492–5500.","mla":"Pentina, Anastasia, et al. Curriculum Learning of Multiple Tasks. IEEE, 2015, pp. 5492–500, doi:10.1109/CVPR.2015.7299188.","chicago":"Pentina, Anastasia, Viktoriia Sharmanska, and Christoph Lampert. “Curriculum Learning of Multiple Tasks,” 5492–5500. IEEE, 2015. https://doi.org/10.1109/CVPR.2015.7299188."},"main_file_link":[{"open_access":"1","url":"http://arxiv.org/abs/1412.1353"}],"page":"5492 - 5500","quality_controlled":"1","doi":"10.1109/CVPR.2015.7299188","date_published":"2015-06-01T00:00:00Z","conference":{"start_date":"2015-06-07","location":"Boston, MA, United States","end_date":"2015-06-12","name":"CVPR: Computer Vision and Pattern Recognition"},"language":[{"iso":"eng"}],"scopus_import":1,"day":"01","month":"06","_id":"1857","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2015","publisher":"IEEE","department":[{"_id":"ChLa"}],"publication_status":"published","status":"public","title":"Curriculum learning of multiple tasks","author":[{"id":"42E87FC6-F248-11E8-B48F-1D18A9856A87","last_name":"Pentina","first_name":"Anastasia","full_name":"Pentina, Anastasia"},{"full_name":"Sharmanska, Viktoriia","orcid":"0000-0003-0192-9308","id":"2EA6D09E-F248-11E8-B48F-1D18A9856A87","last_name":"Sharmanska","first_name":"Viktoriia"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"oa_version":"Preprint","date_created":"2018-12-11T11:54:23Z","date_updated":"2023-02-23T10:17:31Z","type":"conference","publist_id":"5243","abstract":[{"lang":"eng","text":"Sharing information between multiple tasks enables algorithms to achieve good generalization performance even from small amounts of training data. However, in a realistic scenario of multi-task learning not all tasks are equally related to each other, hence it could be advantageous to transfer information only between the most related tasks. In this work we propose an approach that processes multiple tasks in a sequence with sharing between subsequent tasks instead of solving all tasks jointly. Subsequently, we address the question of curriculum learning of tasks, i.e. finding the best order of tasks to be learned. Our approach is based on a generalization bound criterion for choosing the task order that optimizes the average expected classification performance over all tasks. Our experimental results show that learning multiple related tasks sequentially can be more effective than learning them jointly, the order in which tasks are being solved affects the overall performance, and that our model is able to automatically discover the favourable order of tasks. "}]},{"type":"conference","oa_version":"Published Version","file":[{"content_type":"application/pdf","file_size":1674241,"creator":"dernst","file_name":"2015_ECAL_Martius.pdf","access_level":"open_access","date_created":"2023-05-02T07:02:59Z","date_updated":"2023-05-02T07:02:59Z","checksum":"880eabe59c9df12f06a882aa1bc4e600","success":1,"relation":"main_file","file_id":"12882"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"12881","status":"public","ddc":["000"],"title":"Quantifying self-organizing behavior of autonomous robots","day":"01","article_processing_charge":"No","has_accepted_license":"1","scopus_import":"1","date_published":"2015-07-01T00:00:00Z","publication":"Proceedings of the 13th European Conference on Artificial Life","citation":{"mla":"Martius, Georg S., and Eckehard Olbrich. “Quantifying Self-Organizing Behavior of Autonomous Robots.” Proceedings of the 13th European Conference on Artificial Life, MIT Press, 2015, p. 78, doi:10.7551/978-0-262-33027-5-ch018.","short":"G.S. Martius, E. Olbrich, in:, Proceedings of the 13th European Conference on Artificial Life, MIT Press, 2015, p. 78.","chicago":"Martius, Georg S, and Eckehard Olbrich. “Quantifying Self-Organizing Behavior of Autonomous Robots.” In Proceedings of the 13th European Conference on Artificial Life, 78. MIT Press, 2015. https://doi.org/10.7551/978-0-262-33027-5-ch018.","ama":"Martius GS, Olbrich E. Quantifying self-organizing behavior of autonomous robots. In: Proceedings of the 13th European Conference on Artificial Life. MIT Press; 2015:78. doi:10.7551/978-0-262-33027-5-ch018","ista":"Martius GS, Olbrich E. 2015. Quantifying self-organizing behavior of autonomous robots. Proceedings of the 13th European Conference on Artificial Life. ECAL: European Conference on Artificial Life, 78.","ieee":"G. S. Martius and E. Olbrich, “Quantifying self-organizing behavior of autonomous robots,” in Proceedings of the 13th European Conference on Artificial Life, York, United Kingdom, 2015, p. 78.","apa":"Martius, G. S., & Olbrich, E. (2015). Quantifying self-organizing behavior of autonomous robots. In Proceedings of the 13th European Conference on Artificial Life (p. 78). York, United Kingdom: MIT Press. https://doi.org/10.7551/978-0-262-33027-5-ch018"},"page":"78","file_date_updated":"2023-05-02T07:02:59Z","ec_funded":1,"author":[{"id":"3A276B68-F248-11E8-B48F-1D18A9856A87","last_name":"Martius","first_name":"Georg S","full_name":"Martius, Georg S"},{"full_name":"Olbrich, Eckehard","last_name":"Olbrich","first_name":"Eckehard"}],"date_updated":"2023-05-02T07:06:21Z","date_created":"2023-04-30T22:01:07Z","year":"2015","acknowledgement":"This work was supported by the DFG (SPP 1527) and the EU (FP7, REA grant no 291734).","publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"MIT Press","month":"07","publication_identifier":{"isbn":["9780262330275"]},"conference":{"name":"ECAL: European Conference on Artificial Life","end_date":"2015-07-24","location":"York, United Kingdom","start_date":"2015-07-20"},"doi":"10.7551/978-0-262-33027-5-ch018","language":[{"iso":"eng"}],"oa":1,"tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"quality_controlled":"1","project":[{"name":"International IST Postdoc Fellowship Programme","call_identifier":"FP7","grant_number":"291734","_id":"25681D80-B435-11E9-9278-68D0E5697425"}]},{"article_processing_charge":"No","has_accepted_license":"1","day":"01","citation":{"ama":"Sharmanska V. Learning with attributes for object recognition: Parametric and non-parametrics views. 2015. doi:10.15479/at:ista:1401","ieee":"V. Sharmanska, “Learning with attributes for object recognition: Parametric and non-parametrics views,” Institute of Science and Technology Austria, 2015.","apa":"Sharmanska, V. (2015). Learning with attributes for object recognition: Parametric and non-parametrics views. Institute of Science and Technology Austria. https://doi.org/10.15479/at:ista:1401","ista":"Sharmanska V. 2015. Learning with attributes for object recognition: Parametric and non-parametrics views. Institute of Science and Technology Austria.","short":"V. Sharmanska, Learning with Attributes for Object Recognition: Parametric and Non-Parametrics Views, Institute of Science and Technology Austria, 2015.","mla":"Sharmanska, Viktoriia. Learning with Attributes for Object Recognition: Parametric and Non-Parametrics Views. Institute of Science and Technology Austria, 2015, doi:10.15479/at:ista:1401.","chicago":"Sharmanska, Viktoriia. “Learning with Attributes for Object Recognition: Parametric and Non-Parametrics Views.” Institute of Science and Technology Austria, 2015. https://doi.org/10.15479/at:ista:1401."},"page":"144","date_published":"2015-04-01T00:00:00Z","type":"dissertation","alternative_title":["ISTA Thesis"],"abstract":[{"text":"The human ability to recognize objects in complex scenes has driven research in the computer vision field over couple of decades. This thesis focuses on the object recognition task in images. That is, given the image, we want the computer system to be able to predict the class of the object that appears in the image. A recent successful attempt to bridge semantic understanding of the image perceived by humans and by computers uses attribute-based models. Attributes are semantic properties of the objects shared across different categories, which humans and computers can decide on. To explore the attribute-based models we take a statistical machine learning approach, and address two key learning challenges in view of object recognition task: learning augmented attributes as mid-level discriminative feature representation, and learning with attributes as privileged information. Our main contributions are parametric and non-parametric models and algorithms to solve these frameworks. In the parametric approach, we explore an autoencoder model combined with the large margin nearest neighbor principle for mid-level feature learning, and linear support vector machines for learning with privileged information. In the non-parametric approach, we propose a supervised Indian Buffet Process for automatic augmentation of semantic attributes, and explore the Gaussian Processes classification framework for learning with privileged information. A thorough experimental analysis shows the effectiveness of the proposed models in both parametric and non-parametric views.","lang":"eng"}],"_id":"1401","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","ddc":["000"],"status":"public","title":"Learning with attributes for object recognition: Parametric and non-parametrics views","oa_version":"Published Version","file":[{"success":1,"checksum":"3605b402bb6934e09ae4cf672c84baf7","date_updated":"2021-02-22T11:33:17Z","date_created":"2021-02-22T11:33:17Z","file_id":"9177","relation":"main_file","creator":"dernst","content_type":"application/pdf","file_size":7964342,"access_level":"open_access","file_name":"2015_Thesis_Sharmanska.pdf"},{"date_created":"2021-11-16T14:40:45Z","date_updated":"2021-11-17T13:47:24Z","checksum":"e37593b3ee75bf3180629df2d6ca8f4e","file_id":"10297","relation":"main_file","creator":"cchlebak","content_type":"application/pdf","file_size":7372241,"file_name":"2015_Thesis_Sharmanska_pdfa.pdf","access_level":"closed"}],"publication_identifier":{"issn":["2663-337X"]},"month":"04","main_file_link":[{"url":"http://users.sussex.ac.uk/~nq28/viktoriia/Thesis_Sharmanska.pdf"}],"oa":1,"doi":"10.15479/at:ista:1401","language":[{"iso":"eng"}],"supervisor":[{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph"}],"degree_awarded":"PhD","publist_id":"5806","file_date_updated":"2021-11-17T13:47:24Z","year":"2015","acknowledgement":"I would like to thank my supervisor, Christoph Lampert, for guidance throughout my studies and for patience in transforming me into a scientist, and my thesis committee, Chris Wojtan and Horst Bischof, for their help and advice. \r\n\r\nI would like to thank Elisabeth Hacker who perfectly assisted all my administrative needs and was always nice and friendly to me, and the campus team for making the IST Austria campus my second home. \r\nI was honored to collaborate with brilliant researchers and to learn from their experience. Undoubtedly, I learned most of all from Novi Quadrianto: brainstorming our projects and getting exciting results was the most enjoyable part of my work – thank you! I am also grateful to David Knowles, Zoubin Ghahramani, Daniel Hernández-Lobato, Kristian Kersting and Anastasia Pentina for the fantastic projects we worked on together, and to Kristen Grauman and Adriana Kovashka for the exceptional experience working with user studies. I would like to thank my colleagues at IST Austria and my office mates who shared their happy moods, scientific breakthroughs and thought-provoking conversations with me: Chao, Filip, Rustem, Asya, Sameh, Alex, Vlad, Mayu, Neel, Csaba, Thomas, Vladimir, Cristina, Alex Z., Avro, Amelie and Emilie, Andreas H. and Andreas E., Chris, Lena, Michael, Ali and Ipek, Vera, Igor, Katia. Special thanks to Morten for the countless games of table soccer we played together and the tournaments we teamed up for: we will definitely win next time:) A very warm hug to Asya for always being so inspiring and supportive to me, and for helping me to increase the proportion of female computer scientists in our group. ","department":[{"_id":"ChLa"},{"_id":"GradSch"}],"publisher":"Institute of Science and Technology Austria","publication_status":"published","author":[{"full_name":"Sharmanska, Viktoriia","first_name":"Viktoriia","last_name":"Sharmanska","id":"2EA6D09E-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0003-0192-9308"}],"date_created":"2018-12-11T11:51:48Z","date_updated":"2023-09-07T11:40:11Z"},{"month":"10","project":[{"_id":"25681D80-B435-11E9-9278-68D0E5697425","grant_number":"291734","call_identifier":"FP7","name":"International IST Postdoc Fellowship Programme"}],"quality_controlled":"1","oa":1,"tmp":{"name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","short":"CC BY (4.0)","image":"/images/cc_by.png"},"language":[{"iso":"eng"}],"doi":"10.3390/e17107266","ec_funded":1,"publist_id":"5495","file_date_updated":"2020-07-14T12:45:08Z","department":[{"_id":"ChLa"},{"_id":"GaTk"}],"publisher":"MDPI","publication_status":"published","acknowledgement":"This work was supported by the DFG priority program 1527 (Autonomous Learning) and by the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 318723 (MatheMACS) and from the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7/2007-2013) under REA grant agreement no. 291734.","year":"2015","volume":17,"date_created":"2018-12-11T11:53:17Z","date_updated":"2023-10-17T11:42:00Z","author":[{"first_name":"Georg S","last_name":"Martius","id":"3A276B68-F248-11E8-B48F-1D18A9856A87","full_name":"Martius, Georg S"},{"full_name":"Olbrich, Eckehard","first_name":"Eckehard","last_name":"Olbrich"}],"scopus_import":"1","article_processing_charge":"No","has_accepted_license":"1","day":"23","page":"7266 - 7297","citation":{"ama":"Martius GS, Olbrich E. Quantifying emergent behavior of autonomous robots. Entropy. 2015;17(10):7266-7297. doi:10.3390/e17107266","ieee":"G. S. Martius and E. Olbrich, “Quantifying emergent behavior of autonomous robots,” Entropy, vol. 17, no. 10. MDPI, pp. 7266–7297, 2015.","apa":"Martius, G. S., & Olbrich, E. (2015). Quantifying emergent behavior of autonomous robots. Entropy. MDPI. https://doi.org/10.3390/e17107266","ista":"Martius GS, Olbrich E. 2015. Quantifying emergent behavior of autonomous robots. Entropy. 17(10), 7266–7297.","short":"G.S. Martius, E. Olbrich, Entropy 17 (2015) 7266–7297.","mla":"Martius, Georg S., and Eckehard Olbrich. “Quantifying Emergent Behavior of Autonomous Robots.” Entropy, vol. 17, no. 10, MDPI, 2015, pp. 7266–97, doi:10.3390/e17107266.","chicago":"Martius, Georg S, and Eckehard Olbrich. “Quantifying Emergent Behavior of Autonomous Robots.” Entropy. MDPI, 2015. https://doi.org/10.3390/e17107266."},"publication":"Entropy","date_published":"2015-10-23T00:00:00Z","type":"journal_article","issue":"10","abstract":[{"text":"Quantifying behaviors of robots which were generated autonomously from task-independent objective functions is an important prerequisite for objective comparisons of algorithms and movements of animals. The temporal sequence of such a behavior can be considered as a time series and hence complexity measures developed for time series are natural candidates for its quantification. The predictive information and the excess entropy are such complexity measures. They measure the amount of information the past contains about the future and thus quantify the nonrandom structure in the temporal sequence. However, when using these measures for systems with continuous states one has to deal with the fact that their values will depend on the resolution with which the systems states are observed. For deterministic systems both measures will diverge with increasing resolution. We therefore propose a new decomposition of the excess entropy in resolution dependent and resolution independent parts and discuss how they depend on the dimensionality of the dynamics, correlations and the noise level. For the practical estimation we propose to use estimates based on the correlation integral instead of the direct estimation of the mutual information based on next neighbor statistics because the latter allows less control of the scale dependencies. Using our algorithm we are able to show how autonomous learning generates behavior of increasing complexity with increasing learning duration.","lang":"eng"}],"intvolume":" 17","title":"Quantifying emergent behavior of autonomous robots","ddc":["000"],"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"1655","oa_version":"Published Version","file":[{"access_level":"open_access","file_name":"IST-2016-464-v1+1_entropy-17-07266.pdf","content_type":"application/pdf","file_size":6455007,"creator":"system","relation":"main_file","file_id":"4943","checksum":"945d99631a96e0315acb26dc8541dcf9","date_created":"2018-12-12T10:12:25Z","date_updated":"2020-07-14T12:45:08Z"}],"pubrep_id":"464"},{"day":"01","month":"01","scopus_import":1,"series_title":"From Algorithms to Robot Experiments","date_published":"2014-01-01T00:00:00Z","doi":"10.1007/978-3-319-03194-1_3","language":[{"iso":"eng"}],"publication":"Learning Motor Skills","citation":{"chicago":"Muelling, Katharina, Oliver Kroemer, Christoph Lampert, and Bernhard Schölkopf. “Movement Templates for Learning of Hitting and Batting.” In Learning Motor Skills, edited by Jens Kober and Jan Peters, 97:69–82. From Algorithms to Robot Experiments. Springer, 2014. https://doi.org/10.1007/978-3-319-03194-1_3.","mla":"Muelling, Katharina, et al. “Movement Templates for Learning of Hitting and Batting.” Learning Motor Skills, edited by Jens Kober and Jan Peters, vol. 97, Springer, 2014, pp. 69–82, doi:10.1007/978-3-319-03194-1_3.","short":"K. Muelling, O. Kroemer, C. Lampert, B. Schölkopf, in:, J. Kober, J. Peters (Eds.), Learning Motor Skills, Springer, 2014, pp. 69–82.","ista":"Muelling K, Kroemer O, Lampert C, Schölkopf B. 2014.Movement templates for learning of hitting and batting. In: Learning Motor Skills. Springer Tracts in Advanced Robotics, vol. 97, 69–82.","ieee":"K. Muelling, O. Kroemer, C. Lampert, and B. Schölkopf, “Movement templates for learning of hitting and batting,” in Learning Motor Skills, vol. 97, J. Kober and J. Peters, Eds. Springer, 2014, pp. 69–82.","apa":"Muelling, K., Kroemer, O., Lampert, C., & Schölkopf, B. (2014). Movement templates for learning of hitting and batting. In J. Kober & J. Peters (Eds.), Learning Motor Skills (Vol. 97, pp. 69–82). Springer. https://doi.org/10.1007/978-3-319-03194-1_3","ama":"Muelling K, Kroemer O, Lampert C, Schölkopf B. Movement templates for learning of hitting and batting. In: Kober J, Peters J, eds. Learning Motor Skills. Vol 97. From Algorithms to Robot Experiments. Springer; 2014:69-82. doi:10.1007/978-3-319-03194-1_3"},"quality_controlled":"1","page":"69 - 82","abstract":[{"lang":"eng","text":"Hitting and batting tasks, such as tennis forehands, ping-pong strokes, or baseball batting, depend on predictions where the ball can be intercepted and how it can properly be returned to the opponent. These predictions get more accurate over time, hence the behaviors need to be continuously modified. As a result, movement templates with a learned global shape need to be adapted during the execution so that the racket reaches a target position and velocity that will return the ball over to the other side of the net or court. It requires altering learned movements to hit a varying target with the necessary velocity at a specific instant in time. Such a task cannot be incorporated straightforwardly in most movement representations suitable for learning. For example, the standard formulation of the dynamical system based motor primitives (introduced by Ijspeert et al (2002b)) does not satisfy this property despite their flexibility which has allowed learning tasks ranging from locomotion to kendama. In order to fulfill this requirement, we reformulate the Ijspeert framework to incorporate the possibility of specifying a desired hitting point and a desired hitting velocity while maintaining all advantages of the original formulation.We show that the proposed movement template formulation works well in two scenarios, i.e., for hitting a ball on a string with a table tennis racket at a specified velocity and for returning balls launched by a ball gun successfully over the net using forehand movements."}],"publist_id":"5274","type":"book_chapter","alternative_title":["Springer Tracts in Advanced Robotics"],"author":[{"first_name":"Katharina","last_name":"Muelling","full_name":"Muelling, Katharina"},{"full_name":"Kroemer, Oliver","first_name":"Oliver","last_name":"Kroemer"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"},{"full_name":"Schölkopf, Bernhard","first_name":"Bernhard","last_name":"Schölkopf"}],"date_updated":"2021-01-12T06:53:28Z","date_created":"2018-12-11T11:54:14Z","volume":97,"oa_version":"None","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","_id":"1829","year":"2014","title":"Movement templates for learning of hitting and batting","publication_status":"published","status":"public","department":[{"_id":"ChLa"}],"publisher":"Springer","intvolume":" 97","editor":[{"full_name":"Kober, Jens","first_name":"Jens","last_name":"Kober"},{"last_name":"Peters","first_name":"Jan","full_name":"Peters, Jan"}]},{"type":"conference","abstract":[{"text":"The learning with privileged information setting has recently attracted a lot of attention within the machine learning community, as it allows the integration of additional knowledge into the training process of a classifier, even when this comes in the form of a data modality that is not available at test time. Here, we show that privileged information can naturally be treated as noise in the latent function of a Gaussian process classifier (GPC). That is, in contrast to the standard GPC setting, the latent function is not just a nuisance but a feature: it becomes a natural measure of confidence about the training data by modulating the slope of the GPC probit likelihood function. Extensive experiments on public datasets show that the proposed GPC method using privileged noise, called GPC+, improves over a standard GPC without privileged knowledge, and also over the current state-of-the-art SVM-based method, SVM+. Moreover, we show that advanced neural networks and deep learning methods can be compressed as privileged information.","lang":"eng"}],"publist_id":"5038","issue":"January","publication_status":"published","title":"Mind the nuisance: Gaussian process classification using privileged noise","status":"public","department":[{"_id":"ChLa"}],"publisher":"Neural Information Processing Systems","intvolume":" 1","_id":"2033","year":"2014","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","date_updated":"2023-02-23T10:25:24Z","date_created":"2018-12-11T11:55:20Z","oa_version":"Submitted Version","volume":1,"author":[{"full_name":"Hernandez Lobato, Daniel","first_name":"Daniel","last_name":"Hernandez Lobato"},{"last_name":"Sharmanska","first_name":"Viktoriia","orcid":"0000-0003-0192-9308","id":"2EA6D09E-F248-11E8-B48F-1D18A9856A87","full_name":"Sharmanska, Viktoriia"},{"first_name":"Kristian","last_name":"Kersting","full_name":"Kersting, Kristian"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"},{"last_name":"Quadrianto","first_name":"Novi","full_name":"Quadrianto, Novi"}],"scopus_import":1,"month":"12","day":"08","quality_controlled":"1","page":"837-845","publication":"Advances in Neural Information Processing Systems","oa":1,"citation":{"apa":"Hernandez Lobato, D., Sharmanska, V., Kersting, K., Lampert, C., & Quadrianto, N. (2014). Mind the nuisance: Gaussian process classification using privileged noise. In Advances in Neural Information Processing Systems (Vol. 1, pp. 837–845). Montreal, Canada: Neural Information Processing Systems.","ieee":"D. Hernandez Lobato, V. Sharmanska, K. Kersting, C. Lampert, and N. Quadrianto, “Mind the nuisance: Gaussian process classification using privileged noise,” in Advances in Neural Information Processing Systems, Montreal, Canada, 2014, vol. 1, no. January, pp. 837–845.","ista":"Hernandez Lobato D, Sharmanska V, Kersting K, Lampert C, Quadrianto N. 2014. Mind the nuisance: Gaussian process classification using privileged noise. Advances in Neural Information Processing Systems. NIPS: Neural Information Processing Systems vol. 1, 837–845.","ama":"Hernandez Lobato D, Sharmanska V, Kersting K, Lampert C, Quadrianto N. Mind the nuisance: Gaussian process classification using privileged noise. In: Advances in Neural Information Processing Systems. Vol 1. Neural Information Processing Systems; 2014:837-845.","chicago":"Hernandez Lobato, Daniel, Viktoriia Sharmanska, Kristian Kersting, Christoph Lampert, and Novi Quadrianto. “Mind the Nuisance: Gaussian Process Classification Using Privileged Noise.” In Advances in Neural Information Processing Systems, 1:837–45. Neural Information Processing Systems, 2014.","short":"D. Hernandez Lobato, V. Sharmanska, K. Kersting, C. Lampert, N. Quadrianto, in:, Advances in Neural Information Processing Systems, Neural Information Processing Systems, 2014, pp. 837–845.","mla":"Hernandez Lobato, Daniel, et al. “Mind the Nuisance: Gaussian Process Classification Using Privileged Noise.” Advances in Neural Information Processing Systems, vol. 1, no. January, Neural Information Processing Systems, 2014, pp. 837–45."},"main_file_link":[{"open_access":"1","url":"https://papers.nips.cc/paper/5373-mind-the-nuisance-gaussian-process-classification-using-privileged-noise"}],"language":[{"iso":"eng"}],"conference":{"location":"Montreal, Canada","start_date":"2014-12-08","end_date":"2014-12-13","name":"NIPS: Neural Information Processing Systems"},"date_published":"2014-12-08T00:00:00Z"},{"scopus_import":1,"day":"01","publication":"Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)","citation":{"ieee":"E. Morvant, A. Habrard, and S. Ayache, “Majority vote of diverse classifiers for late fusion,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Joensuu, Finland, 2014, vol. 8621, pp. 153–162.","apa":"Morvant, E., Habrard, A., & Ayache, S. (2014). Majority vote of diverse classifiers for late fusion. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8621, pp. 153–162). Joensuu, Finland: Springer. https://doi.org/10.1007/978-3-662-44415-3_16","ista":"Morvant E, Habrard A, Ayache S. 2014. Majority vote of diverse classifiers for late fusion. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). IAPR: International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, LNCS, vol. 8621, 153–162.","ama":"Morvant E, Habrard A, Ayache S. Majority vote of diverse classifiers for late fusion. In: Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol 8621. Springer; 2014:153-162. doi:10.1007/978-3-662-44415-3_16","chicago":"Morvant, Emilie, Amaury Habrard, and Stéphane Ayache. “Majority Vote of Diverse Classifiers for Late Fusion.” In Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8621:153–62. Springer, 2014. https://doi.org/10.1007/978-3-662-44415-3_16.","short":"E. Morvant, A. Habrard, S. Ayache, in:, Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer, 2014, pp. 153–162.","mla":"Morvant, Emilie, et al. “Majority Vote of Diverse Classifiers for Late Fusion.” Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8621, Springer, 2014, pp. 153–62, doi:10.1007/978-3-662-44415-3_16."},"page":"153 - 162","date_published":"2014-01-01T00:00:00Z","type":"conference","alternative_title":["LNCS"],"abstract":[{"lang":"eng","text":"In the past few years, a lot of attention has been devoted to multimedia indexing by fusing multimodal informations. Two kinds of fusion schemes are generally considered: The early fusion and the late fusion. We focus on late classifier fusion, where one combines the scores of each modality at the decision level. To tackle this problem, we investigate a recent and elegant well-founded quadratic program named MinCq coming from the machine learning PAC-Bayesian theory. MinCq looks for the weighted combination, over a set of real-valued functions seen as voters, leading to the lowest misclassification rate, while maximizing the voters’ diversity. We propose an extension of MinCq tailored to multimedia indexing. Our method is based on an order-preserving pairwise loss adapted to ranking that allows us to improve Mean Averaged Precision measure while taking into account the diversity of the voters that we want to fuse. We provide evidence that this method is naturally adapted to late fusion procedures and confirm the good behavior of our approach on the challenging PASCAL VOC’07 benchmark."}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"2057","status":"public","title":"Majority vote of diverse classifiers for late fusion","intvolume":" 8621","oa_version":"Preprint","month":"01","external_id":{"arxiv":["1404.7796"]},"oa":1,"main_file_link":[{"url":"http://arxiv.org/abs/1404.7796","open_access":"1"}],"quality_controlled":"1","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"conference":{"end_date":"2014-08-22","location":"Joensuu, Finland","start_date":"2014-08-20","name":"IAPR: International Workshop on Structural, Syntactic, and Statistical Pattern Recognition"},"doi":"10.1007/978-3-662-44415-3_16","language":[{"iso":"eng"}],"ec_funded":1,"publist_id":"4989","year":"2014","publication_status":"published","publisher":"Springer","department":[{"_id":"ChLa"}],"author":[{"full_name":"Morvant, Emilie","first_name":"Emilie","last_name":"Morvant","id":"4BAC2A72-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8301-7240"},{"last_name":"Habrard","first_name":"Amaury","full_name":"Habrard, Amaury"},{"last_name":"Ayache","first_name":"Stéphane","full_name":"Ayache, Stéphane"}],"date_created":"2018-12-11T11:55:28Z","date_updated":"2021-01-12T06:55:01Z","volume":8621},{"ec_funded":1,"publist_id":"4813","author":[{"first_name":"Alexander","last_name":"Kolesnikov","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87","full_name":"Kolesnikov, Alexander"},{"full_name":"Guillaumin, Matthieu","last_name":"Guillaumin","first_name":"Matthieu"},{"first_name":"Vittorio","last_name":"Ferrari","full_name":"Ferrari, Vittorio"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"volume":8691,"date_created":"2018-12-11T11:56:07Z","date_updated":"2021-01-12T06:55:46Z","year":"2014","editor":[{"full_name":"Fleet, David","first_name":"David","last_name":"Fleet"},{"full_name":"Pajdla, Tomas","last_name":"Pajdla","first_name":"Tomas"},{"last_name":"Schiele","first_name":"Bernt","full_name":"Schiele, Bernt"},{"full_name":"Tuytelaars, Tinne","last_name":"Tuytelaars","first_name":"Tinne"}],"department":[{"_id":"ChLa"}],"publisher":"Springer","publication_status":"published","month":"09","doi":"10.1007/978-3-319-10578-9_36","conference":{"location":"Zurich, Switzerland","start_date":"2014-09-06","end_date":"2014-09-12","name":"ECCV: European Conference on Computer Vision"},"language":[{"iso":"eng"}],"main_file_link":[{"url":"http://arxiv.org/abs/1403.7057","open_access":"1"}],"oa":1,"project":[{"grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425","name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7"}],"quality_controlled":"1","issue":"PART 3","abstract":[{"text":"We present LS-CRF, a new method for training cyclic Conditional Random Fields (CRFs) from large datasets that is inspired by classical closed-form expressions for the maximum likelihood parameters of a generative graphical model with tree topology. Training a CRF with LS-CRF requires only solving a set of independent regression problems, each of which can be solved efficiently in closed form or by an iterative solver. This makes LS-CRF orders of magnitude faster than classical CRF training based on probabilistic inference, and at the same time more flexible and easier to implement than other approximate techniques, such as pseudolikelihood or piecewise training. We apply LS-CRF to the task of semantic image segmentation, showing that it achieves on par accuracy to other training techniques at higher speed, thereby allowing efficient CRF training from very large training sets. For example, training a linearly parameterized pairwise CRF on 150,000 images requires less than one hour on a modern workstation.","lang":"eng"}],"type":"conference","alternative_title":["LNCS"],"oa_version":"Submitted Version","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","_id":"2171","intvolume":" 8691","title":"Closed-form approximate CRF training for scalable image segmentation","status":"public","day":"01","scopus_import":1,"date_published":"2014-09-01T00:00:00Z","citation":{"ama":"Kolesnikov A, Guillaumin M, Ferrari V, Lampert C. Closed-form approximate CRF training for scalable image segmentation. In: Fleet D, Pajdla T, Schiele B, Tuytelaars T, eds. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol 8691. Springer; 2014:550-565. doi:10.1007/978-3-319-10578-9_36","ieee":"A. Kolesnikov, M. Guillaumin, V. Ferrari, and C. Lampert, “Closed-form approximate CRF training for scalable image segmentation,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Zurich, Switzerland, 2014, vol. 8691, no. PART 3, pp. 550–565.","apa":"Kolesnikov, A., Guillaumin, M., Ferrari, V., & Lampert, C. (2014). Closed-form approximate CRF training for scalable image segmentation. In D. Fleet, T. Pajdla, B. Schiele, & T. Tuytelaars (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8691, pp. 550–565). Zurich, Switzerland: Springer. https://doi.org/10.1007/978-3-319-10578-9_36","ista":"Kolesnikov A, Guillaumin M, Ferrari V, Lampert C. 2014. Closed-form approximate CRF training for scalable image segmentation. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). ECCV: European Conference on Computer Vision, LNCS, vol. 8691, 550–565.","short":"A. Kolesnikov, M. Guillaumin, V. Ferrari, C. Lampert, in:, D. Fleet, T. Pajdla, B. Schiele, T. Tuytelaars (Eds.), Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer, 2014, pp. 550–565.","mla":"Kolesnikov, Alexander, et al. “Closed-Form Approximate CRF Training for Scalable Image Segmentation.” Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), edited by David Fleet et al., vol. 8691, no. PART 3, Springer, 2014, pp. 550–65, doi:10.1007/978-3-319-10578-9_36.","chicago":"Kolesnikov, Alexander, Matthieu Guillaumin, Vittorio Ferrari, and Christoph Lampert. “Closed-Form Approximate CRF Training for Scalable Image Segmentation.” In Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), edited by David Fleet, Tomas Pajdla, Bernt Schiele, and Tinne Tuytelaars, 8691:550–65. Springer, 2014. https://doi.org/10.1007/978-3-319-10578-9_36."},"publication":"Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)","page":"550 - 565"},{"abstract":[{"lang":"eng","text":"In this work we introduce a new approach to co-classification, i.e. the task of jointly classifying multiple, otherwise independent, data samples. The method we present, named CoConut, is based on the idea of adding a regularizer in the label space to encode certain priors on the resulting labelings. A regularizer that encourages labelings that are smooth across the test set, for instance, can be seen as a test-time variant of the cluster assumption, which has been proven useful at training time in semi-supervised learning. A regularizer that introduces a preference for certain class proportions can be regarded as a prior distribution on the class labels. CoConut can build on existing classifiers without making any assumptions on how they were obtained and without the need to re-train them. The use of a regularizer adds a new level of flexibility. It allows the integration of potentially new information at test time, even in other modalities than what the classifiers were trained on. We evaluate our framework on six datasets, reporting a clear performance gain in classification accuracy compared to the standard classification setup that predicts labels for each test sample separately.\r\n"}],"type":"conference","oa_version":"Published Version","file":[{"relation":"main_file","file_id":"4683","checksum":"c4c6d3efdb8ee648faf3e76849839ce2","date_updated":"2020-07-14T12:45:31Z","date_created":"2018-12-12T10:08:23Z","access_level":"open_access","file_name":"IST-2016-490-v1+1_khamis-bmvc2014.pdf","file_size":408172,"content_type":"application/pdf","creator":"system"}],"pubrep_id":"490","title":"CoConut: Co-classification with output space regularization","ddc":["000"],"status":"public","_id":"2173","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","day":"01","has_accepted_license":"1","scopus_import":1,"date_published":"2014-09-01T00:00:00Z","publication":"Proceedings of the British Machine Vision Conference 2014","citation":{"mla":"Khamis, Sameh, and Christoph Lampert. “CoConut: Co-Classification with Output Space Regularization.” Proceedings of the British Machine Vision Conference 2014, BMVA Press, 2014.","short":"S. Khamis, C. Lampert, in:, Proceedings of the British Machine Vision Conference 2014, BMVA Press, 2014.","chicago":"Khamis, Sameh, and Christoph Lampert. “CoConut: Co-Classification with Output Space Regularization.” In Proceedings of the British Machine Vision Conference 2014. BMVA Press, 2014.","ama":"Khamis S, Lampert C. CoConut: Co-classification with output space regularization. In: Proceedings of the British Machine Vision Conference 2014. BMVA Press; 2014.","ista":"Khamis S, Lampert C. 2014. CoConut: Co-classification with output space regularization. Proceedings of the British Machine Vision Conference 2014. BMVC: British Machine Vision Conference.","ieee":"S. Khamis and C. Lampert, “CoConut: Co-classification with output space regularization,” in Proceedings of the British Machine Vision Conference 2014, Nottingham, UK, 2014.","apa":"Khamis, S., & Lampert, C. (2014). CoConut: Co-classification with output space regularization. In Proceedings of the British Machine Vision Conference 2014. Nottingham, UK: BMVA Press."},"file_date_updated":"2020-07-14T12:45:31Z","ec_funded":1,"publist_id":"4811","date_updated":"2021-01-12T06:55:46Z","date_created":"2018-12-11T11:56:08Z","author":[{"last_name":"Khamis","first_name":"Sameh","full_name":"Khamis, Sameh"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"publication_status":"published","publisher":"BMVA Press","department":[{"_id":"ChLa"}],"year":"2014","month":"09","language":[{"iso":"eng"}],"conference":{"name":"BMVC: British Machine Vision Conference","end_date":"2014-09-05","start_date":"2014-09-01","location":"Nottingham, UK"},"quality_controlled":"1","project":[{"grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425","call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding"}],"oa":1},{"month":"09","day":"24","scopus_import":1,"language":[{"iso":"eng"}],"conference":{"name":"CVPR: Computer Vision and Pattern Recognition","start_date":"2014-06-23","location":"Columbus, USA","end_date":"2014-06-28"},"doi":"10.1109/CVPR.2014.182","date_published":"2014-09-24T00:00:00Z","quality_controlled":"1","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"page":"1402 - 1409","publication":"Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition","citation":{"short":"V. Sydorov, M. Sakurada, C. Lampert, in:, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE, 2014, pp. 1402–1409.","mla":"Sydorov, Vladyslav, et al. “Deep Fisher Kernels – End to End Learning of the Fisher Kernel GMM Parameters.” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE, 2014, pp. 1402–09, doi:10.1109/CVPR.2014.182.","chicago":"Sydorov, Vladyslav, Mayu Sakurada, and Christoph Lampert. “Deep Fisher Kernels – End to End Learning of the Fisher Kernel GMM Parameters.” In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1402–9. IEEE, 2014. https://doi.org/10.1109/CVPR.2014.182.","ama":"Sydorov V, Sakurada M, Lampert C. Deep Fisher Kernels – End to end learning of the Fisher Kernel GMM parameters. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE; 2014:1402-1409. doi:10.1109/CVPR.2014.182","ieee":"V. Sydorov, M. Sakurada, and C. Lampert, “Deep Fisher Kernels – End to end learning of the Fisher Kernel GMM parameters,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Columbus, USA, 2014, pp. 1402–1409.","apa":"Sydorov, V., Sakurada, M., & Lampert, C. (2014). Deep Fisher Kernels – End to end learning of the Fisher Kernel GMM parameters. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (pp. 1402–1409). Columbus, USA: IEEE. https://doi.org/10.1109/CVPR.2014.182","ista":"Sydorov V, Sakurada M, Lampert C. 2014. Deep Fisher Kernels – End to end learning of the Fisher Kernel GMM parameters. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR: Computer Vision and Pattern Recognition, 1402–1409."},"abstract":[{"text":"Fisher Kernels and Deep Learning were two developments with significant impact on large-scale object categorization in the last years. Both approaches were shown to achieve state-of-the-art results on large-scale object categorization datasets, such as ImageNet. Conceptually, however, they are perceived as very different and it is not uncommon for heated debates to spring up when advocates of both paradigms meet at conferences or workshops. In this work, we emphasize the similarities between both architectures rather than their differences and we argue that such a unified view allows us to transfer ideas from one domain to the other. As a concrete example we introduce a method for learning a support vector machine classifier with Fisher kernel at the same time as a task-specific data representation. We reinterpret the setting as a multi-layer feed forward network. Its final layer is the classifier, parameterized by a weight vector, and the two previous layers compute Fisher vectors, parameterized by the coefficients of a Gaussian mixture model. We introduce a gradient descent based learning algorithm that, in contrast to other feature learning techniques, is not just derived from intuition or biological analogy, but has a theoretical justification in the framework of statistical learning theory. Our experiments show that the new training procedure leads to significant improvements in classification accuracy while preserving the modularity and geometric interpretability of a support vector machine setup.","lang":"eng"}],"ec_funded":1,"publist_id":"4812","type":"conference","date_created":"2018-12-11T11:56:08Z","date_updated":"2021-01-12T06:55:46Z","oa_version":"None","author":[{"full_name":"Sydorov, Vladyslav","first_name":"Vladyslav","last_name":"Sydorov"},{"full_name":"Sakurada, Mayu","last_name":"Sakurada","first_name":"Mayu"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"publication_status":"published","title":"Deep Fisher Kernels – End to end learning of the Fisher Kernel GMM parameters","status":"public","publisher":"IEEE","department":[{"_id":"ChLa"}],"user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","_id":"2172","year":"2014"},{"publist_id":"4802","ec_funded":1,"department":[{"_id":"ChLa"}],"publisher":"Springer","publication_status":"published","year":"2014","acknowledgement":"This work was funded by the French project SoLSTiCe ANR-13-BS02-01 of the ANR. ","volume":97,"date_created":"2018-12-11T11:56:10Z","date_updated":"2021-01-12T06:55:49Z","author":[{"first_name":"Aurélien","last_name":"Bellet","full_name":"Bellet, Aurélien"},{"last_name":"Habrard","first_name":"Amaury","full_name":"Habrard, Amaury"},{"id":"4BAC2A72-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8301-7240","first_name":"Emilie","last_name":"Morvant","full_name":"Morvant, Emilie"},{"last_name":"Sebban","first_name":"Marc","full_name":"Sebban, Marc"}],"month":"10","project":[{"name":"Lifelong Learning of Visual Scene Understanding","call_identifier":"FP7","_id":"2532554C-B435-11E9-9278-68D0E5697425","grant_number":"308036"}],"quality_controlled":"1","main_file_link":[{"open_access":"1","url":"https://hal.archives-ouvertes.fr/hal-01009578/document"}],"oa":1,"language":[{"iso":"eng"}],"doi":"10.1007/s10994-014-5462-z","type":"journal_article","issue":"1-2","abstract":[{"lang":"eng","text":"Weighted majority votes allow one to combine the output of several classifiers or voters. MinCq is a recent algorithm for optimizing the weight of each voter based on the minimization of a theoretical bound over the risk of the vote with elegant PAC-Bayesian generalization guarantees. However, while it has demonstrated good performance when combining weak classifiers, MinCq cannot make use of the useful a priori knowledge that one may have when using a mixture of weak and strong voters. In this paper, we propose P-MinCq, an extension of MinCq that can incorporate such knowledge in the form of a constraint over the distribution of the weights, along with general proofs of convergence that stand in the sample compression setting for data-dependent voters. The approach is applied to a vote of k-NN classifiers with a specific modeling of the voters' performance. P-MinCq significantly outperforms the classic k-NN classifier, a symmetric NN and MinCq using the same voters. We show that it is also competitive with LMNN, a popular metric learning algorithm, and that combining both approaches further reduces the error."}],"intvolume":" 97","title":"Learning a priori constrained weighted majority votes","status":"public","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","_id":"2180","oa_version":"Submitted Version","scopus_import":1,"day":"01","page":"129 - 154","citation":{"ama":"Bellet A, Habrard A, Morvant E, Sebban M. Learning a priori constrained weighted majority votes. Machine Learning. 2014;97(1-2):129-154. doi:10.1007/s10994-014-5462-z","apa":"Bellet, A., Habrard, A., Morvant, E., & Sebban, M. (2014). Learning a priori constrained weighted majority votes. Machine Learning. Springer. https://doi.org/10.1007/s10994-014-5462-z","ieee":"A. Bellet, A. Habrard, E. Morvant, and M. Sebban, “Learning a priori constrained weighted majority votes,” Machine Learning, vol. 97, no. 1–2. Springer, pp. 129–154, 2014.","ista":"Bellet A, Habrard A, Morvant E, Sebban M. 2014. Learning a priori constrained weighted majority votes. Machine Learning. 97(1–2), 129–154.","short":"A. Bellet, A. Habrard, E. Morvant, M. Sebban, Machine Learning 97 (2014) 129–154.","mla":"Bellet, Aurélien, et al. “Learning a Priori Constrained Weighted Majority Votes.” Machine Learning, vol. 97, no. 1–2, Springer, 2014, pp. 129–54, doi:10.1007/s10994-014-5462-z.","chicago":"Bellet, Aurélien, Amaury Habrard, Emilie Morvant, and Marc Sebban. “Learning a Priori Constrained Weighted Majority Votes.” Machine Learning. Springer, 2014. https://doi.org/10.1007/s10994-014-5462-z."},"publication":"Machine Learning","date_published":"2014-10-01T00:00:00Z"},{"publisher":"Elsevier","department":[{"_id":"ChLa"}],"intvolume":" 1","title":"Adaptation de domaine de vote de majorité par auto-étiquetage non itératif","status":"public","publication_status":"published","_id":"2189","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2014","oa_version":"Preprint","volume":1,"date_created":"2018-12-11T11:56:13Z","date_updated":"2021-01-12T06:55:52Z","author":[{"full_name":"Morvant, Emilie","id":"4BAC2A72-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8301-7240","first_name":"Emilie","last_name":"Morvant"}],"type":"conference","publist_id":"4785","abstract":[{"text":"En apprentissage automatique, nous parlons d'adaptation de domaine lorsque les données de test (cibles) et d'apprentissage (sources) sont générées selon différentes distributions. Nous devons donc développer des algorithmes de classification capables de s'adapter à une nouvelle distribution, pour laquelle aucune information sur les étiquettes n'est disponible. Nous attaquons cette problématique sous l'angle de l'approche PAC-Bayésienne qui se focalise sur l'apprentissage de modèles définis comme des votes de majorité sur un ensemble de fonctions. Dans ce contexte, nous introduisons PV-MinCq une version adaptative de l'algorithme (non adaptatif) MinCq. PV-MinCq suit le principe suivant. Nous transférons les étiquettes sources aux points cibles proches pour ensuite appliquer MinCq sur l'échantillon cible ``auto-étiqueté'' (justifié par une borne théorique). Plus précisément, nous définissons un auto-étiquetage non itératif qui se focalise dans les régions où les distributions marginales source et cible sont les plus similaires. Dans un second temps, nous étudions l'influence de notre auto-étiquetage pour en déduire une procédure de validation des hyperparamètres. Finalement, notre approche montre des résultats empiriques prometteurs.","lang":"fre"}],"page":"49-58","quality_controlled":"1","main_file_link":[{"url":"https://hal.archives-ouvertes.fr/hal-01005776/","open_access":"1"}],"citation":{"mla":"Morvant, Emilie. Adaptation de Domaine de Vote de Majorité Par Auto-Étiquetage Non Itératif. Vol. 1, Elsevier, 2014, pp. 49–58.","short":"E. Morvant, in:, Elsevier, 2014, pp. 49–58.","chicago":"Morvant, Emilie. “Adaptation de Domaine de Vote de Majorité Par Auto-Étiquetage Non Itératif,” 1:49–58. Elsevier, 2014.","ama":"Morvant E. Adaptation de domaine de vote de majorité par auto-étiquetage non itératif. In: Vol 1. Elsevier; 2014:49-58.","ista":"Morvant E. 2014. Adaptation de domaine de vote de majorité par auto-étiquetage non itératif. CAP: Conférence Francophone sur l’Apprentissage Automatique (Machine Learning French Conference) vol. 1, 49–58.","ieee":"E. Morvant, “Adaptation de domaine de vote de majorité par auto-étiquetage non itératif,” presented at the CAP: Conférence Francophone sur l’Apprentissage Automatique (Machine Learning French Conference), Saint-Etienne, France, 2014, vol. 1, pp. 49–58.","apa":"Morvant, E. (2014). Adaptation de domaine de vote de majorité par auto-étiquetage non itératif (Vol. 1, pp. 49–58). Presented at the CAP: Conférence Francophone sur l’Apprentissage Automatique (Machine Learning French Conference), Saint-Etienne, France: Elsevier."},"oa":1,"language":[{"iso":"eng"}],"date_published":"2014-07-01T00:00:00Z","conference":{"name":"CAP: Conférence Francophone sur l'Apprentissage Automatique (Machine Learning French Conference)","location":"Saint-Etienne, France"},"article_processing_charge":"No","month":"07","day":"01"},{"language":[{"iso":"eng"}],"conference":{"start_date":"2014-06-21","location":"Beijing, China","end_date":"2014-06-26","name":"ICML: International Conference on Machine Learning"},"date_published":"2014-05-10T00:00:00Z","quality_controlled":"1","page":"991 - 999","main_file_link":[{"open_access":"1","url":"https://dl.acm.org/citation.cfm?id=3045003"}],"citation":{"chicago":"Pentina, Anastasia, and Christoph Lampert. “A PAC-Bayesian Bound for Lifelong Learning,” 32:991–99. ML Research Press, 2014.","mla":"Pentina, Anastasia, and Christoph Lampert. A PAC-Bayesian Bound for Lifelong Learning. Vol. 32, ML Research Press, 2014, pp. 991–99.","short":"A. Pentina, C. Lampert, in:, ML Research Press, 2014, pp. 991–999.","ista":"Pentina A, Lampert C. 2014. A PAC-Bayesian bound for Lifelong Learning. ICML: International Conference on Machine Learning vol. 32, 991–999.","ieee":"A. Pentina and C. Lampert, “A PAC-Bayesian bound for Lifelong Learning,” presented at the ICML: International Conference on Machine Learning, Beijing, China, 2014, vol. 32, pp. 991–999.","apa":"Pentina, A., & Lampert, C. (2014). A PAC-Bayesian bound for Lifelong Learning (Vol. 32, pp. 991–999). Presented at the ICML: International Conference on Machine Learning, Beijing, China: ML Research Press.","ama":"Pentina A, Lampert C. A PAC-Bayesian bound for Lifelong Learning. In: Vol 32. ML Research Press; 2014:991-999."},"oa":1,"month":"05","day":"10","article_processing_charge":"No","scopus_import":"1","date_created":"2018-12-11T11:56:03Z","date_updated":"2023-10-17T11:54:24Z","oa_version":"Submitted Version","volume":32,"author":[{"full_name":"Pentina, Anastasia","last_name":"Pentina","first_name":"Anastasia","id":"42E87FC6-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}],"title":"A PAC-Bayesian bound for Lifelong Learning","status":"public","publication_status":"published","intvolume":" 32","publisher":"ML Research Press","department":[{"_id":"ChLa"}],"year":"2014","_id":"2160","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"text":"Transfer learning has received a lot of attention in the machine learning community over the last years, and several effective algorithms have been developed. However, relatively little is known about their theoretical properties, especially in the setting of lifelong learning, where the goal is to transfer information to tasks for which no data have been observed so far. In this work we study lifelong learning from a theoretical perspective. Our main result is a PAC-Bayesian generalization bound that offers a unified view on existing paradigms for transfer learning, such as the transfer of parameters or the transfer of low-dimensional representations. We also use the bound to derive two principled lifelong learning algorithms, and we show that these yield results comparable with existing methods.","lang":"eng"}],"publist_id":"4844","type":"conference"},{"doi":"10.1109/ICCV.2013.139","date_published":"2013-12-01T00:00:00Z","conference":{"location":"Sydney, Australia","start_date":"2013-12-01","end_date":"2013-12-08","name":"ICCV: International Conference on Computer Vision"},"language":[{"iso":"eng"}],"main_file_link":[{"open_access":"1","url":"http://www.cv-foundation.org/openaccess/ICCV2013.py"}],"oa":1,"citation":{"mla":"Kazmar, Tomas, et al. Drosophila Embryo Stage Annotation Using Label Propagation. IEEE, 2013, doi:10.1109/ICCV.2013.139.","short":"T. Kazmar, E. Kvon, A. Stark, C. Lampert, in:, IEEE, 2013.","chicago":"Kazmar, Tomas, Evgeny Kvon, Alexander Stark, and Christoph Lampert. “Drosophila Embryo Stage Annotation Using Label Propagation.” IEEE, 2013. https://doi.org/10.1109/ICCV.2013.139.","ama":"Kazmar T, Kvon E, Stark A, Lampert C. Drosophila Embryo Stage Annotation using Label Propagation. In: IEEE; 2013. doi:10.1109/ICCV.2013.139","ista":"Kazmar T, Kvon E, Stark A, Lampert C. 2013. Drosophila Embryo Stage Annotation using Label Propagation. ICCV: International Conference on Computer Vision.","ieee":"T. Kazmar, E. Kvon, A. Stark, and C. Lampert, “Drosophila Embryo Stage Annotation using Label Propagation,” presented at the ICCV: International Conference on Computer Vision, Sydney, Australia, 2013.","apa":"Kazmar, T., Kvon, E., Stark, A., & Lampert, C. (2013). Drosophila Embryo Stage Annotation using Label Propagation. Presented at the ICCV: International Conference on Computer Vision, Sydney, Australia: IEEE. https://doi.org/10.1109/ICCV.2013.139"},"project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"quality_controlled":"1","month":"12","day":"01","scopus_import":1,"author":[{"first_name":"Tomas","last_name":"Kazmar","full_name":"Kazmar, Tomas"},{"last_name":"Kvon","first_name":"Evgeny","full_name":"Kvon, Evgeny"},{"full_name":"Stark, Alexander","last_name":"Stark","first_name":"Alexander"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"oa_version":"Submitted Version","date_created":"2018-12-11T11:56:49Z","date_updated":"2021-01-12T06:56:35Z","year":"2013","_id":"2294","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publisher":"IEEE","department":[{"_id":"ChLa"}],"publication_status":"published","status":"public","title":"Drosophila Embryo Stage Annotation using Label Propagation","ec_funded":1,"publist_id":"4634","abstract":[{"lang":"eng","text":"In this work we propose a system for automatic classification of Drosophila embryos into developmental stages.\r\nWhile the system is designed to solve an actual problem in biological research, we believe that the principle underly-\r\ning it is interesting not only for biologists, but also for researchers in computer vision. The main idea is to combine two orthogonal sources of information: one is a classifier trained on strongly invariant features, which makes it applicable to images of very different conditions, but also leads to rather noisy predictions. The other is a label propagation step based on a more powerful similarity measure that however is only consistent within specific subsets of the data at a time.\r\nIn our biological setup, the information sources are the shape and the staining patterns of embryo images. We show\r\nexperimentally that while neither of the methods can be used by itself to achieve satisfactory results, their combina-\r\ntion achieves prediction quality comparable to human performance."}],"type":"conference"},{"day":"01","month":"12","scopus_import":1,"conference":{"name":"ICCV: International Conference on Computer Vision","end_date":"2013-12-08","start_date":"2013-12-01","location":"Sydney, Australia"},"doi":"10.1109/ICCV.2013.107","date_published":"2013-12-01T00:00:00Z","language":[{"iso":"eng"}],"main_file_link":[{"url":"www.cv-foundation.org/openaccess/content_iccv_2013/papers/Sharmanska_Learning_to_Rank_2013_ICCV_paper.pdf","open_access":"1"}],"citation":{"short":"V. Sharmanska, N. Quadrianto, C. Lampert, in:, IEEE, 2013, pp. 825–832.","mla":"Sharmanska, Viktoriia, et al. Learning to Rank Using Privileged Information. IEEE, 2013, pp. 825–32, doi:10.1109/ICCV.2013.107.","chicago":"Sharmanska, Viktoriia, Novi Quadrianto, and Christoph Lampert. “Learning to Rank Using Privileged Information,” 825–32. IEEE, 2013. https://doi.org/10.1109/ICCV.2013.107.","ama":"Sharmanska V, Quadrianto N, Lampert C. Learning to rank using privileged information. In: IEEE; 2013:825-832. doi:10.1109/ICCV.2013.107","apa":"Sharmanska, V., Quadrianto, N., & Lampert, C. (2013). Learning to rank using privileged information (pp. 825–832). Presented at the ICCV: International Conference on Computer Vision, Sydney, Australia: IEEE. https://doi.org/10.1109/ICCV.2013.107","ieee":"V. Sharmanska, N. Quadrianto, and C. Lampert, “Learning to rank using privileged information,” presented at the ICCV: International Conference on Computer Vision, Sydney, Australia, 2013, pp. 825–832.","ista":"Sharmanska V, Quadrianto N, Lampert C. 2013. Learning to rank using privileged information. ICCV: International Conference on Computer Vision, 825–832."},"oa":1,"quality_controlled":"1","project":[{"call_identifier":"FP7","name":"Lifelong Learning of Visual Scene Understanding","grant_number":"308036","_id":"2532554C-B435-11E9-9278-68D0E5697425"}],"page":"825 - 832","abstract":[{"lang":"eng","text":"Many computer vision problems have an asymmetric distribution of information between training and test time. In this work, we study the case where we are given additional information about the training data, which however will not be available at test time. This situation is called learning using privileged information (LUPI). We introduce two maximum-margin techniques that are able to make use of this additional source of information, and we show that the framework is applicable to several scenarios that have been studied in computer vision before. Experiments with attributes, bounding boxes, image tags and rationales as additional information in object classification show promising results."}],"ec_funded":1,"publist_id":"4635","type":"conference","author":[{"orcid":"0000-0003-0192-9308","id":"2EA6D09E-F248-11E8-B48F-1D18A9856A87","last_name":"Sharmanska","first_name":"Viktoriia","full_name":"Sharmanska, Viktoriia"},{"full_name":"Quadrianto, Novi","last_name":"Quadrianto","first_name":"Novi"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"date_created":"2018-12-11T11:56:49Z","date_updated":"2023-02-23T10:36:41Z","oa_version":"Submitted Version","_id":"2293","year":"2013","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication_status":"published","title":"Learning to rank using privileged information","status":"public","publisher":"IEEE","department":[{"_id":"ChLa"}]},{"language":[{"iso":"eng"}],"date_published":"2013-07-30T00:00:00Z","doi":"10.1109/TPAMI.2013.140","quality_controlled":"1","page":"453 - 465","publication":"IEEE Transactions on Pattern Analysis and Machine Intelligence","citation":{"ama":"Lampert C, Nickisch H, Harmeling S. Attribute-based classification for zero-shot learning of object categories. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2013;36(3):453-465. doi:10.1109/TPAMI.2013.140","apa":"Lampert, C., Nickisch, H., & Harmeling, S. (2013). Attribute-based classification for zero-shot learning of object categories. IEEE Transactions on Pattern Analysis and Machine Intelligence. IEEE. https://doi.org/10.1109/TPAMI.2013.140","ieee":"C. Lampert, H. Nickisch, and S. Harmeling, “Attribute-based classification for zero-shot learning of object categories,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 3. IEEE, pp. 453–465, 2013.","ista":"Lampert C, Nickisch H, Harmeling S. 2013. Attribute-based classification for zero-shot learning of object categories. IEEE Transactions on Pattern Analysis and Machine Intelligence. 36(3), 453–465.","short":"C. Lampert, H. Nickisch, S. Harmeling, IEEE Transactions on Pattern Analysis and Machine Intelligence 36 (2013) 453–465.","mla":"Lampert, Christoph, et al. “Attribute-Based Classification for Zero-Shot Learning of Object Categories.” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 3, IEEE, 2013, pp. 453–65, doi:10.1109/TPAMI.2013.140.","chicago":"Lampert, Christoph, Hannes Nickisch, and Stefan Harmeling. “Attribute-Based Classification for Zero-Shot Learning of Object Categories.” IEEE Transactions on Pattern Analysis and Machine Intelligence. IEEE, 2013. https://doi.org/10.1109/TPAMI.2013.140."},"month":"07","day":"30","scopus_import":1,"date_updated":"2021-01-12T06:57:58Z","date_created":"2018-12-11T11:58:08Z","volume":36,"oa_version":"None","author":[{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"},{"first_name":"Hannes","last_name":"Nickisch","full_name":"Nickisch, Hannes"},{"full_name":"Harmeling, Stefan","last_name":"Harmeling","first_name":"Stefan"}],"title":"Attribute-based classification for zero-shot learning of object categories","publication_status":"published","status":"public","intvolume":" 36","department":[{"_id":"ChLa"}],"publisher":"IEEE","year":"2013","_id":"2516","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"text":"We study the problem of object recognition for categories for which we have no training examples, a task also called zero-data or zero-shot learning. This situation has hardly been studied in computer vision research, even though it occurs frequently: the world contains tens of thousands of different object classes and for only few of them image collections have been formed and suitably annotated. To tackle the problem we introduce attribute-based classification: objects are identified based on a high-level description that is phrased in terms of semantic attributes, such as the object's color or shape. Because the identification of each such property transcends the specific learning task at hand, the attribute classifiers can be pre-learned independently, e.g. from existing image datasets unrelated to the current task. Afterwards, new classes can be detected based on their attribute representation, without the need for a new training phase. In this paper we also introduce a new dataset, Animals with Attributes, of over 30,000 images of 50 animal classes, annotated with 85 semantic attributes. Extensive experiments on this and two more datasets show that attribute-based classification indeed is able to categorize images without access to any training images of the target classes.","lang":"eng"}],"publist_id":"4385","issue":"3","type":"journal_article"},{"quality_controlled":"1","oa":1,"language":[{"iso":"eng"}],"conference":{"name":"UAI: Uncertainty in Artificial Intelligence","end_date":"2013-07-15","start_date":"2013-07-11","location":"Bellevue, WA, United States"},"month":"07","publication_identifier":{"isbn":["9780974903996"]},"publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"AUAI Press","year":"2013","date_created":"2018-12-11T11:58:09Z","date_updated":"2023-02-23T10:46:36Z","author":[{"last_name":"Quadrianto","first_name":"Novi","full_name":"Quadrianto, Novi"},{"full_name":"Sharmanska, Viktoriia","orcid":"0000-0003-0192-9308","id":"2EA6D09E-F248-11E8-B48F-1D18A9856A87","last_name":"Sharmanska","first_name":"Viktoriia"},{"full_name":"Knowles, David","last_name":"Knowles","first_name":"David"},{"first_name":"Zoubin","last_name":"Ghahramani","full_name":"Ghahramani, Zoubin"}],"file_date_updated":"2020-07-14T12:45:42Z","publist_id":"4381","page":"527 - 536","publication":"Proceedings of the 29th conference uncertainty in Artificial Intelligence","citation":{"ieee":"N. Quadrianto, V. Sharmanska, D. Knowles, and Z. Ghahramani, “The supervised IBP: Neighbourhood preserving infinite latent feature models,” in Proceedings of the 29th conference uncertainty in Artificial Intelligence, Bellevue, WA, United States, 2013, pp. 527–536.","apa":"Quadrianto, N., Sharmanska, V., Knowles, D., & Ghahramani, Z. (2013). The supervised IBP: Neighbourhood preserving infinite latent feature models. In Proceedings of the 29th conference uncertainty in Artificial Intelligence (pp. 527–536). Bellevue, WA, United States: AUAI Press.","ista":"Quadrianto N, Sharmanska V, Knowles D, Ghahramani Z. 2013. The supervised IBP: Neighbourhood preserving infinite latent feature models. Proceedings of the 29th conference uncertainty in Artificial Intelligence. UAI: Uncertainty in Artificial Intelligence, 527–536.","ama":"Quadrianto N, Sharmanska V, Knowles D, Ghahramani Z. The supervised IBP: Neighbourhood preserving infinite latent feature models. In: Proceedings of the 29th Conference Uncertainty in Artificial Intelligence. AUAI Press; 2013:527-536.","chicago":"Quadrianto, Novi, Viktoriia Sharmanska, David Knowles, and Zoubin Ghahramani. “The Supervised IBP: Neighbourhood Preserving Infinite Latent Feature Models.” In Proceedings of the 29th Conference Uncertainty in Artificial Intelligence, 527–36. AUAI Press, 2013.","short":"N. Quadrianto, V. Sharmanska, D. Knowles, Z. Ghahramani, in:, Proceedings of the 29th Conference Uncertainty in Artificial Intelligence, AUAI Press, 2013, pp. 527–536.","mla":"Quadrianto, Novi, et al. “The Supervised IBP: Neighbourhood Preserving Infinite Latent Feature Models.” Proceedings of the 29th Conference Uncertainty in Artificial Intelligence, AUAI Press, 2013, pp. 527–36."},"date_published":"2013-07-11T00:00:00Z","scopus_import":1,"day":"11","has_accepted_license":"1","ddc":["000"],"status":"public","title":"The supervised IBP: Neighbourhood preserving infinite latent feature models","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"2520","file":[{"relation":"main_file","file_id":"5134","checksum":"325f20c4b926bd74d39006b97df572bd","date_updated":"2020-07-14T12:45:42Z","date_created":"2018-12-12T10:15:16Z","access_level":"open_access","file_name":"IST-2013-137-v1+1_QuaShaKnoGha13.pdf","file_size":1117100,"content_type":"application/pdf","creator":"system"}],"oa_version":"Submitted Version","pubrep_id":"137","type":"conference","abstract":[{"text":"We propose a probabilistic model to infer supervised latent variables in\r\nthe Hamming space from observed data. Our model allows simultaneous\r\ninference of the number of binary latent variables, and their values. The\r\nlatent variables preserve neighbourhood structure of the data in a sense\r\nthat objects in the same semantic concept have similar latent values, and\r\nobjects in different concepts have dissimilar latent values. We formulate\r\nthe supervised infinite latent variable problem based on an intuitive\r\nprinciple of pulling objects together if they are of the same type, and\r\npushing them apart if they are not. We then combine this principle with a\r\nflexible Indian Buffet Process prior on the latent variables. We show that\r\nthe inferred supervised latent variables can be directly used to perform a\r\nnearest neighbour search for the purpose of retrieval. We introduce a new\r\napplication of dynamically extending hash codes, and show how to\r\neffectively couple the structure of the hash codes with continuously\r\ngrowing structure of the neighbourhood preserving infinite latent feature\r\nspace.","lang":"eng"}]},{"month":"01","day":"01","scopus_import":1,"language":[{"iso":"eng"}],"conference":{"name":" AISTATS: Conference on Uncertainty in Artificial Intelligence","location":"Scottsdale, AZ, United States","start_date":"2013-04-29","end_date":"2013-05-01"},"date_published":"2013-01-01T00:00:00Z","quality_controlled":"1","page":"161 - 169","main_file_link":[{"url":"http://jmlr.org/proceedings/papers/v31/chen13a.html","open_access":"1"}],"oa":1,"citation":{"ista":"Chen C, Kolmogorov V, Yan Z, Metaxas D, Lampert C. 2013. Computing the M most probable modes of a graphical model. AISTATS: Conference on Uncertainty in Artificial Intelligence, JMLR: W&CP, vol. 31, 161–169.","apa":"Chen, C., Kolmogorov, V., Yan, Z., Metaxas, D., & Lampert, C. (2013). Computing the M most probable modes of a graphical model (Vol. 31, pp. 161–169). Presented at the AISTATS: Conference on Uncertainty in Artificial Intelligence, Scottsdale, AZ, United States: JMLR.","ieee":"C. Chen, V. Kolmogorov, Z. Yan, D. Metaxas, and C. Lampert, “Computing the M most probable modes of a graphical model,” presented at the AISTATS: Conference on Uncertainty in Artificial Intelligence, Scottsdale, AZ, United States, 2013, vol. 31, pp. 161–169.","ama":"Chen C, Kolmogorov V, Yan Z, Metaxas D, Lampert C. Computing the M most probable modes of a graphical model. In: Vol 31. JMLR; 2013:161-169.","chicago":"Chen, Chao, Vladimir Kolmogorov, Zhu Yan, Dimitris Metaxas, and Christoph Lampert. “Computing the M Most Probable Modes of a Graphical Model,” 31:161–69. JMLR, 2013.","mla":"Chen, Chao, et al. Computing the M Most Probable Modes of a Graphical Model. Vol. 31, JMLR, 2013, pp. 161–69.","short":"C. Chen, V. Kolmogorov, Z. Yan, D. Metaxas, C. Lampert, in:, JMLR, 2013, pp. 161–169."},"abstract":[{"lang":"eng","text":" We introduce the M-modes problem for graphical models: predicting the M label configurations of highest probability that are at the same time local maxima of the probability landscape. M-modes have multiple possible applications: because they are intrinsically diverse, they provide a principled alternative to non-maximum suppression techniques for structured prediction, they can act as codebook vectors for quantizing the configuration space, or they can form component centers for mixture model approximation. We present two algorithms for solving the M-modes problem. The first algorithm solves the problem in polynomial time when the underlying graphical model is a simple chain. The second algorithm solves the problem for junction chains. In synthetic and real dataset, we demonstrate how M-modes can improve the performance of prediction. We also use the generated modes as a tool to understand the topography of the probability distribution of configurations, for example with relation to the training set size and amount of noise in the data. "}],"publist_id":"3846","alternative_title":[" JMLR: W&CP"],"type":"conference","date_updated":"2021-01-12T07:00:35Z","date_created":"2018-12-11T12:00:14Z","oa_version":"None","volume":31,"author":[{"id":"3E92416E-F248-11E8-B48F-1D18A9856A87","first_name":"Chao","last_name":"Chen","full_name":"Chen, Chao"},{"full_name":"Kolmogorov, Vladimir","first_name":"Vladimir","last_name":"Kolmogorov","id":"3D50B0BA-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Yan","first_name":"Zhu","full_name":"Yan, Zhu"},{"full_name":"Metaxas, Dimitris","last_name":"Metaxas","first_name":"Dimitris"},{"orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph"}],"status":"public","publication_status":"published","title":"Computing the M most probable modes of a graphical model","department":[{"_id":"HeEd"},{"_id":"VlKo"},{"_id":"ChLa"}],"intvolume":" 31","publisher":"JMLR","_id":"2901","year":"2013","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87"},{"intvolume":" 7724","status":"public","title":"Beyond dataset bias: Multi-task unaligned shared knowledge transfer","ddc":["000"],"_id":"2948","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","file":[{"file_id":"5874","relation":"main_file","checksum":"a0a7234a89e2192af655b0d0ae3bf445","date_created":"2019-01-22T14:03:11Z","date_updated":"2020-07-14T12:45:55Z","access_level":"open_access","file_name":"2012_ACCV_Tommasi.pdf","creator":"dernst","content_type":"application/pdf","file_size":1513620}],"oa_version":"Submitted Version","alternative_title":["LNCS"],"type":"conference","abstract":[{"text":"Many visual datasets are traditionally used to analyze the performance of different learning techniques. The evaluation is usually done within each dataset, therefore it is questionable if such results are a reliable indicator of true generalization ability. We propose here an algorithm to exploit the existing data resources when learning on a new multiclass problem. Our main idea is to identify an image representation that decomposes orthogonally into two subspaces: a part specific to each dataset, and a part generic to, and therefore shared between, all the considered source sets. This allows us to use the generic representation as un-biased reference knowledge for a novel classification task. By casting the method in the multi-view setting, we also make it possible to use different features for different databases. We call the algorithm MUST, Multitask Unaligned Shared knowledge Transfer. Through extensive experiments on five public datasets, we show that MUST consistently improves the cross-datasets generalization performance.","lang":"eng"}],"page":"1 - 15","citation":{"short":"T. Tommasi, N. Quadrianto, B. Caputo, C. Lampert, 7724 (2013) 1–15.","mla":"Tommasi, Tatiana, et al. Beyond Dataset Bias: Multi-Task Unaligned Shared Knowledge Transfer. Vol. 7724, Springer, 2013, pp. 1–15, doi:10.1007/978-3-642-37331-2_1.","chicago":"Tommasi, Tatiana, Novi Quadrianto, Barbara Caputo, and Christoph Lampert. “Beyond Dataset Bias: Multi-Task Unaligned Shared Knowledge Transfer.” Lecture Notes in Computer Science. Springer, 2013. https://doi.org/10.1007/978-3-642-37331-2_1.","ama":"Tommasi T, Quadrianto N, Caputo B, Lampert C. Beyond dataset bias: Multi-task unaligned shared knowledge transfer. 2013;7724:1-15. doi:10.1007/978-3-642-37331-2_1","ieee":"T. Tommasi, N. Quadrianto, B. Caputo, and C. Lampert, “Beyond dataset bias: Multi-task unaligned shared knowledge transfer,” vol. 7724. Springer, pp. 1–15, 2013.","apa":"Tommasi, T., Quadrianto, N., Caputo, B., & Lampert, C. (2013). Beyond dataset bias: Multi-task unaligned shared knowledge transfer. Presented at the ACCV: Asian Conference on Computer Vision, Daejeon, Korea: Springer. https://doi.org/10.1007/978-3-642-37331-2_1","ista":"Tommasi T, Quadrianto N, Caputo B, Lampert C. 2013. Beyond dataset bias: Multi-task unaligned shared knowledge transfer. 7724, 1–15."},"date_published":"2013-04-04T00:00:00Z","series_title":"Lecture Notes in Computer Science","scopus_import":1,"has_accepted_license":"1","day":"04","publisher":"Springer","department":[{"_id":"ChLa"}],"publication_status":"published","year":"2013","acknowledgement":"This work was supported by the PASCAL 2 Network of Excellence (TT) and by the Newton International Fellowship (NQ)","volume":7724,"date_created":"2018-12-11T12:00:30Z","date_updated":"2020-08-11T10:09:54Z","author":[{"full_name":"Tommasi, Tatiana","first_name":"Tatiana","last_name":"Tommasi"},{"last_name":"Quadrianto","first_name":"Novi","full_name":"Quadrianto, Novi"},{"first_name":"Barbara","last_name":"Caputo","full_name":"Caputo, Barbara"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"publist_id":"3784","file_date_updated":"2020-07-14T12:45:55Z","quality_controlled":"1","oa":1,"language":[{"iso":"eng"}],"doi":"10.1007/978-3-642-37331-2_1","conference":{"location":"Daejeon, Korea","start_date":"2012-11-05","end_date":"2012-11-09","name":"ACCV: Asian Conference on Computer Vision"},"month":"04"},{"author":[{"last_name":"Quadrianto","first_name":"Novi","full_name":"Quadrianto, Novi"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"volume":3,"oa_version":"None","date_updated":"2021-01-12T07:42:38Z","date_created":"2018-12-11T12:02:39Z","_id":"3321","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2013","intvolume":" 3","editor":[{"first_name":"Werner","last_name":"Dubitzky","full_name":"Dubitzky, Werner"},{"full_name":"Wolkenhauer, Olaf","first_name":"Olaf","last_name":"Wolkenhauer"},{"full_name":"Cho, Kwang","first_name":"Kwang","last_name":"Cho"},{"full_name":"Yokota, Hiroki","last_name":"Yokota","first_name":"Hiroki"}],"department":[{"_id":"ChLa"}],"publisher":"Springer","status":"public","publication_status":"published","title":"Kernel based learning","publist_id":"3314","type":"encyclopedia_article","doi":"10.1007/978-1-4419-9863-7_604","date_published":"2013-01-01T00:00:00Z","language":[{"iso":"eng"}],"citation":{"ama":"Quadrianto N, Lampert C. Kernel based learning. In: Dubitzky W, Wolkenhauer O, Cho K, Yokota H, eds. Encyclopedia of Systems Biology. Vol 3. Springer; 2013:1069-1069. doi:10.1007/978-1-4419-9863-7_604","ista":"Quadrianto N, Lampert C. 2013.Kernel based learning. In: Encyclopedia of Systems Biology. vol. 3, 1069–1069.","ieee":"N. Quadrianto and C. Lampert, “Kernel based learning,” in Encyclopedia of Systems Biology, vol. 3, W. Dubitzky, O. Wolkenhauer, K. Cho, and H. Yokota, Eds. Springer, 2013, pp. 1069–1069.","apa":"Quadrianto, N., & Lampert, C. (2013). Kernel based learning. In W. Dubitzky, O. Wolkenhauer, K. Cho, & H. Yokota (Eds.), Encyclopedia of Systems Biology (Vol. 3, pp. 1069–1069). Springer. https://doi.org/10.1007/978-1-4419-9863-7_604","mla":"Quadrianto, Novi, and Christoph Lampert. “Kernel Based Learning.” Encyclopedia of Systems Biology, edited by Werner Dubitzky et al., vol. 3, Springer, 2013, pp. 1069–1069, doi:10.1007/978-1-4419-9863-7_604.","short":"N. Quadrianto, C. Lampert, in:, W. Dubitzky, O. Wolkenhauer, K. Cho, H. Yokota (Eds.), Encyclopedia of Systems Biology, Springer, 2013, pp. 1069–1069.","chicago":"Quadrianto, Novi, and Christoph Lampert. “Kernel Based Learning.” In Encyclopedia of Systems Biology, edited by Werner Dubitzky, Olaf Wolkenhauer, Kwang Cho, and Hiroki Yokota, 3:1069–1069. Springer, 2013. https://doi.org/10.1007/978-1-4419-9863-7_604."},"publication":"Encyclopedia of Systems Biology","page":"1069 - 1069","quality_controlled":"1","day":"01","month":"01"},{"citation":{"ama":"Lampert C. Dynamic pruning of factor graphs for maximum marginal prediction. In: Vol 1. Neural Information Processing Systems; 2012:82-90.","ista":"Lampert C. 2012. Dynamic pruning of factor graphs for maximum marginal prediction. NIPS: Neural Information Processing Systems vol. 1, 82–90.","apa":"Lampert, C. (2012). Dynamic pruning of factor graphs for maximum marginal prediction (Vol. 1, pp. 82–90). Presented at the NIPS: Neural Information Processing Systems, Lake Tahoe, NV, United States: Neural Information Processing Systems.","ieee":"C. Lampert, “Dynamic pruning of factor graphs for maximum marginal prediction,” presented at the NIPS: Neural Information Processing Systems, Lake Tahoe, NV, United States, 2012, vol. 1, pp. 82–90.","mla":"Lampert, Christoph. Dynamic Pruning of Factor Graphs for Maximum Marginal Prediction. Vol. 1, Neural Information Processing Systems, 2012, pp. 82–90.","short":"C. Lampert, in:, Neural Information Processing Systems, 2012, pp. 82–90.","chicago":"Lampert, Christoph. “Dynamic Pruning of Factor Graphs for Maximum Marginal Prediction,” 1:82–90. Neural Information Processing Systems, 2012."},"page":"82 - 90","quality_controlled":"1","date_published":"2012-12-01T00:00:00Z","conference":{"name":"NIPS: Neural Information Processing Systems","location":"Lake Tahoe, NV, United States","start_date":"2012-12-03","end_date":"2012-12-06"},"language":[{"iso":"eng"}],"scopus_import":1,"day":"01","month":"12","_id":"2825","year":"2012","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","publisher":"Neural Information Processing Systems","department":[{"_id":"ChLa"}],"intvolume":" 1","title":"Dynamic pruning of factor graphs for maximum marginal prediction","status":"public","publication_status":"published","author":[{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"oa_version":"None","volume":1,"date_updated":"2021-01-12T06:59:59Z","date_created":"2018-12-11T11:59:48Z","type":"conference","publist_id":"3975","abstract":[{"text":"We study the problem of maximum marginal prediction (MMP) in probabilistic graphical models, a task that occurs, for example, as the Bayes optimal decision rule under a Hamming loss. MMP is typically performed as a two-stage procedure: one estimates each variable's marginal probability and then forms a prediction from the states of maximal probability. In this work we propose a simple yet effective technique for accelerating MMP when inference is sampling-based: instead of the above two-stage procedure we directly estimate the posterior probability of each decision variable. This allows us to identify the point of time when we are sufficiently certain about any individual decision. Whenever this is the case, we dynamically prune the variables we are confident about from the underlying factor graph. Consequently, at any time only samples of variables whose decision is still uncertain need to be created. Experiments in two prototypical scenarios, multi-label classification and image inpainting, show that adaptive sampling can drastically accelerate MMP without sacrificing prediction accuracy.","lang":"eng"}]},{"scopus_import":1,"day":"01","month":"09","publication":"International Journal of Computer Vision","citation":{"chicago":"Blaschko, Matthew, and Christoph Lampert. “Guest Editorial: Special Issue on Structured Prediction and Inference.” International Journal of Computer Vision. Springer, 2012. https://doi.org/10.1007/s11263-012-0530-y.","mla":"Blaschko, Matthew, and Christoph Lampert. “Guest Editorial: Special Issue on Structured Prediction and Inference.” International Journal of Computer Vision, vol. 99, no. 3, Springer, 2012, pp. 257–58, doi:10.1007/s11263-012-0530-y.","short":"M. Blaschko, C. Lampert, International Journal of Computer Vision 99 (2012) 257–258.","ista":"Blaschko M, Lampert C. 2012. Guest editorial: Special issue on structured prediction and inference. International Journal of Computer Vision. 99(3), 257–258.","apa":"Blaschko, M., & Lampert, C. (2012). Guest editorial: Special issue on structured prediction and inference. International Journal of Computer Vision. Springer. https://doi.org/10.1007/s11263-012-0530-y","ieee":"M. Blaschko and C. Lampert, “Guest editorial: Special issue on structured prediction and inference,” International Journal of Computer Vision, vol. 99, no. 3. Springer, pp. 257–258, 2012.","ama":"Blaschko M, Lampert C. Guest editorial: Special issue on structured prediction and inference. International Journal of Computer Vision. 2012;99(3):257-258. doi:10.1007/s11263-012-0530-y"},"quality_controlled":"1","page":"257 - 258","doi":"10.1007/s11263-012-0530-y","date_published":"2012-09-01T00:00:00Z","language":[{"iso":"eng"}],"type":"journal_article","abstract":[{"text":"Overview of the Special Issue on structured prediction and inference.","lang":"eng"}],"issue":"3","publist_id":"3521","_id":"3164","year":"2012","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","status":"public","publication_status":"published","title":"Guest editorial: Special issue on structured prediction and inference","department":[{"_id":"ChLa"}],"publisher":"Springer","intvolume":" 99","author":[{"first_name":"Matthew","last_name":"Blaschko","full_name":"Blaschko, Matthew"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"date_created":"2018-12-11T12:01:46Z","date_updated":"2021-01-12T07:41:30Z","oa_version":"None","volume":99},{"article_processing_charge":"No","has_accepted_license":"1","day":"01","scopus_import":1,"date_published":"2012-10-01T00:00:00Z","page":"242 - 255","citation":{"ista":"Sharmanska V, Quadrianto N, Lampert C. 2012. Augmented attribute representations. ECCV: European Conference on Computer Vision, LNCS, vol. 7576, 242–255.","apa":"Sharmanska, V., Quadrianto, N., & Lampert, C. (2012). Augmented attribute representations (Vol. 7576, pp. 242–255). Presented at the ECCV: European Conference on Computer Vision, Florence, Italy: Springer. https://doi.org/10.1007/978-3-642-33715-4_18","ieee":"V. Sharmanska, N. Quadrianto, and C. Lampert, “Augmented attribute representations,” presented at the ECCV: European Conference on Computer Vision, Florence, Italy, 2012, vol. 7576, no. PART 5, pp. 242–255.","ama":"Sharmanska V, Quadrianto N, Lampert C. Augmented attribute representations. In: Vol 7576. Springer; 2012:242-255. doi:10.1007/978-3-642-33715-4_18","chicago":"Sharmanska, Viktoriia, Novi Quadrianto, and Christoph Lampert. “Augmented Attribute Representations,” 7576:242–55. Springer, 2012. https://doi.org/10.1007/978-3-642-33715-4_18.","mla":"Sharmanska, Viktoriia, et al. Augmented Attribute Representations. Vol. 7576, no. PART 5, Springer, 2012, pp. 242–55, doi:10.1007/978-3-642-33715-4_18.","short":"V. Sharmanska, N. Quadrianto, C. Lampert, in:, Springer, 2012, pp. 242–255."},"issue":"PART 5","abstract":[{"text":"We propose a new learning method to infer a mid-level feature representation that combines the advantage of semantic attribute representations with the higher expressive power of non-semantic features. The idea lies in augmenting an existing attribute-based representation with additional dimensions for which an autoencoder model is coupled with a large-margin principle. This construction allows a smooth transition between the zero-shot regime with no training example, the unsupervised regime with training examples but without class labels, and the supervised regime with training examples and with class labels. The resulting optimization problem can be solved efficiently, because several of the necessity steps have closed-form solutions. Through extensive experiments we show that the augmented representation achieves better results in terms of object categorization accuracy than the semantic representation alone.","lang":"eng"}],"alternative_title":["LNCS"],"type":"conference","oa_version":"Submitted Version","file":[{"creator":"dernst","content_type":"application/pdf","file_size":6073897,"file_name":"2012_ECCV_Sharmanska.pdf","access_level":"open_access","date_updated":"2020-07-14T12:46:00Z","date_created":"2020-05-15T12:29:04Z","checksum":"bccdbe0663780d25a1e0524002b2d896","file_id":"7861","relation":"main_file"}],"intvolume":" 7576","status":"public","title":"Augmented attribute representations","ddc":["000"],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"3125","month":"10","language":[{"iso":"eng"}],"doi":"10.1007/978-3-642-33715-4_18","conference":{"name":"ECCV: European Conference on Computer Vision","end_date":"2012-10-13","start_date":"2012-10-07","location":"Florence, Italy"},"quality_controlled":"1","oa":1,"publist_id":"3574","file_date_updated":"2020-07-14T12:46:00Z","volume":7576,"date_updated":"2023-02-23T11:13:25Z","date_created":"2018-12-11T12:01:32Z","author":[{"orcid":"0000-0003-0192-9308","id":"2EA6D09E-F248-11E8-B48F-1D18A9856A87","last_name":"Sharmanska","first_name":"Viktoriia","full_name":"Sharmanska, Viktoriia"},{"first_name":"Novi","last_name":"Quadrianto","full_name":"Quadrianto, Novi"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"}],"department":[{"_id":"ChLa"}],"publisher":"Springer","publication_status":"published","year":"2012"},{"alternative_title":["LNCS"],"type":"conference","publist_id":"3573","abstract":[{"lang":"eng","text":"In this work we propose a new information-theoretic clustering algorithm that infers cluster memberships by direct optimization of a non-parametric mutual information estimate between data distribution and cluster assignment. Although the optimization objective has a solid theoretical foundation it is hard to optimize. We propose an approximate optimization formulation that leads to an efficient algorithm with low runtime complexity. The algorithm has a single free parameter, the number of clusters to find. We demonstrate superior performance on several synthetic and real datasets.\r\n"}],"intvolume":" 7476","publisher":"Springer","department":[{"_id":"ChLa"}],"publication_status":"published","title":"Information theoretic clustering using minimal spanning trees","status":"public","_id":"3126","year":"2012","user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","volume":7476,"oa_version":"None","date_created":"2018-12-11T12:01:32Z","date_updated":"2021-01-12T07:41:14Z","author":[{"full_name":"Müller, Andreas","last_name":"Müller","first_name":"Andreas"},{"first_name":"Sebastian","last_name":"Nowozin","full_name":"Nowozin, Sebastian"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"}],"scopus_import":1,"day":"14","month":"08","page":"205 - 215","quality_controlled":"1","citation":{"chicago":"Müller, Andreas, Sebastian Nowozin, and Christoph Lampert. “Information Theoretic Clustering Using Minimal Spanning Trees,” 7476:205–15. Springer, 2012. https://doi.org/10.1007/978-3-642-32717-9_21.","short":"A. Müller, S. Nowozin, C. Lampert, in:, Springer, 2012, pp. 205–215.","mla":"Müller, Andreas, et al. Information Theoretic Clustering Using Minimal Spanning Trees. Vol. 7476, Springer, 2012, pp. 205–15, doi:10.1007/978-3-642-32717-9_21.","apa":"Müller, A., Nowozin, S., & Lampert, C. (2012). Information theoretic clustering using minimal spanning trees (Vol. 7476, pp. 205–215). Presented at the DAGM: German Association For Pattern Recognition, Graz, Austria: Springer. https://doi.org/10.1007/978-3-642-32717-9_21","ieee":"A. Müller, S. Nowozin, and C. Lampert, “Information theoretic clustering using minimal spanning trees,” presented at the DAGM: German Association For Pattern Recognition, Graz, Austria, 2012, vol. 7476, pp. 205–215.","ista":"Müller A, Nowozin S, Lampert C. 2012. Information theoretic clustering using minimal spanning trees. DAGM: German Association For Pattern Recognition, LNCS, vol. 7476, 205–215.","ama":"Müller A, Nowozin S, Lampert C. Information theoretic clustering using minimal spanning trees. In: Vol 7476. Springer; 2012:205-215. doi:10.1007/978-3-642-32717-9_21"},"language":[{"iso":"eng"}],"date_published":"2012-08-14T00:00:00Z","doi":"10.1007/978-3-642-32717-9_21","conference":{"name":"DAGM: German Association For Pattern Recognition","end_date":"2012-08-31","start_date":"2012-08-28","location":"Graz, Austria"}},{"month":"03","publication_identifier":{"issn":["1861-8200"],"eissn":["1861-8219"]},"doi":"10.1007/s11554-010-0168-3","language":[{"iso":"eng"}],"oa":1,"quality_controlled":"1","file_date_updated":"2020-07-14T12:46:04Z","publist_id":"3417","author":[{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"},{"last_name":"Peters","first_name":"Jan","full_name":"Peters, Jan"}],"date_created":"2018-12-11T12:02:15Z","date_updated":"2022-05-24T08:05:40Z","volume":7,"year":"2012","publication_status":"published","department":[{"_id":"ChLa"}],"publisher":"Springer","day":"01","article_processing_charge":"No","has_accepted_license":"1","scopus_import":"1","date_published":"2012-03-01T00:00:00Z","publication":"Journal of Real-Time Image Processing","citation":{"ieee":"C. Lampert and J. Peters, “Real-time detection of colored objects in multiple camera streams with off-the-shelf hardware components,” Journal of Real-Time Image Processing, vol. 7, no. 1. Springer, pp. 31–41, 2012.","apa":"Lampert, C., & Peters, J. (2012). Real-time detection of colored objects in multiple camera streams with off-the-shelf hardware components. Journal of Real-Time Image Processing. Springer. https://doi.org/10.1007/s11554-010-0168-3","ista":"Lampert C, Peters J. 2012. Real-time detection of colored objects in multiple camera streams with off-the-shelf hardware components. Journal of Real-Time Image Processing. 7(1), 31–41.","ama":"Lampert C, Peters J. Real-time detection of colored objects in multiple camera streams with off-the-shelf hardware components. Journal of Real-Time Image Processing. 2012;7(1):31-41. doi:10.1007/s11554-010-0168-3","chicago":"Lampert, Christoph, and Jan Peters. “Real-Time Detection of Colored Objects in Multiple Camera Streams with off-the-Shelf Hardware Components.” Journal of Real-Time Image Processing. Springer, 2012. https://doi.org/10.1007/s11554-010-0168-3.","short":"C. Lampert, J. Peters, Journal of Real-Time Image Processing 7 (2012) 31–41.","mla":"Lampert, Christoph, and Jan Peters. “Real-Time Detection of Colored Objects in Multiple Camera Streams with off-the-Shelf Hardware Components.” Journal of Real-Time Image Processing, vol. 7, no. 1, Springer, 2012, pp. 31–41, doi:10.1007/s11554-010-0168-3."},"article_type":"original","page":"31 - 41","abstract":[{"lang":"eng","text":"We describe RTblob, a high speed vision system that detects objects in cluttered scenes based on their color and shape at a speed of over 800 frames/s. Because the system is available as open-source software and relies only on off-the-shelf PC hardware components, it can provide the basis for multiple application scenarios. As an illustrative example, we show how RTblob can be used in a robotic table tennis scenario to estimate ball trajectories through 3D space simultaneously from four cameras images at a speed of 200 Hz."}],"issue":"1","type":"journal_article","file":[{"file_name":"2012_Springer_Lampert.pdf","access_level":"open_access","creator":"kschuh","content_type":"application/pdf","file_size":2933187,"file_id":"5958","relation":"main_file","date_updated":"2020-07-14T12:46:04Z","date_created":"2019-02-12T10:52:25Z","checksum":"241be47ea50e81a283bcf4c45b07e8cc"}],"oa_version":"Submitted Version","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"3248","title":"Real-time detection of colored objects in multiple camera streams with off-the-shelf hardware components","ddc":["000"],"status":"public","intvolume":" 7"},{"quality_controlled":"1","citation":{"chicago":"Korc, Filip, Vladimir Kolmogorov, and Christoph Lampert. “Approximating Marginals Using Discrete Energy Minimization.” ICML, 2012.","short":"F. Korc, V. Kolmogorov, C. Lampert, in:, ICML, 2012.","mla":"Korc, Filip, et al. Approximating Marginals Using Discrete Energy Minimization. ICML, 2012.","apa":"Korc, F., Kolmogorov, V., & Lampert, C. (2012). Approximating marginals using discrete energy minimization. Presented at the ICML: International Conference on Machine Learning, Edinburgh, Scotland: ICML.","ieee":"F. Korc, V. Kolmogorov, and C. Lampert, “Approximating marginals using discrete energy minimization,” presented at the ICML: International Conference on Machine Learning, Edinburgh, Scotland, 2012.","ista":"Korc F, Kolmogorov V, Lampert C. 2012. Approximating marginals using discrete energy minimization. ICML: International Conference on Machine Learning, Inferning 2012, .","ama":"Korc F, Kolmogorov V, Lampert C. Approximating marginals using discrete energy minimization. In: ICML; 2012."},"oa":1,"language":[{"iso":"eng"}],"conference":{"name":"ICML: International Conference on Machine Learning","start_date":"2012-06-26","location":"Edinburgh, Scotland","end_date":"2012-07-01"},"date_published":"2012-06-30T00:00:00Z","day":"30","month":"06","has_accepted_license":"1","status":"public","ddc":["000"],"title":"Approximating marginals using discrete energy minimization","publication_status":"published","publisher":"ICML","department":[{"_id":"ChLa"},{"_id":"VlKo"}],"user_id":"3E5EF7F0-F248-11E8-B48F-1D18A9856A87","_id":"3124","year":"2012","date_created":"2018-12-11T12:01:31Z","date_updated":"2023-02-23T12:24:24Z","file":[{"file_size":305836,"content_type":"application/pdf","creator":"system","access_level":"open_access","file_name":"IST-2016-565-v1+1_DM-inferning2012.pdf","checksum":"3d0d4246548c736857302aadb2ff5d15","date_created":"2018-12-12T10:11:34Z","date_updated":"2020-07-14T12:46:00Z","relation":"main_file","file_id":"4889"}],"oa_version":"Submitted Version","author":[{"full_name":"Korc, Filip","id":"476A2FD6-F248-11E8-B48F-1D18A9856A87","first_name":"Filip","last_name":"Korc"},{"last_name":"Kolmogorov","first_name":"Vladimir","id":"3D50B0BA-F248-11E8-B48F-1D18A9856A87","full_name":"Kolmogorov, Vladimir"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"pubrep_id":"565","related_material":{"record":[{"id":"5396","status":"public","relation":"later_version"}]},"alternative_title":["Inferning 2012"],"type":"conference","file_date_updated":"2020-07-14T12:46:00Z","abstract":[{"text":"We consider the problem of inference in a graphical model with binary variables. While in theory it is arguably preferable to compute marginal probabilities, in practice researchers often use MAP inference due to the availability of efficient discrete optimization algorithms. We bridge the gap between the two approaches by introducing the Discrete Marginals technique in which approximate marginals are obtained by minimizing an objective function with unary and pairwise terms over a discretized domain. This allows the use of techniques originally developed for MAP-MRF inference and learning. We explore two ways to set up the objective function - by discretizing the Bethe free energy and by learning it from training data. Experimental results show that for certain types of graphs a learned function can outperform the Bethe approximation. We also establish a link between the Bethe free energy and submodular functions.\r\n","lang":"eng"}],"publist_id":"3575"},{"file_date_updated":"2020-07-14T12:46:44Z","abstract":[{"text":"We consider the problem of inference in agraphical model with binary variables. While in theory it is arguably preferable to compute marginal probabilities, in practice researchers often use MAP inference due to the availability of efficient discrete optimization algorithms. We bridge the gap between the two approaches by introducing the Discrete Marginals technique in which approximate marginals are obtained by minimizing an objective function with unary and pair-wise terms over a discretized domain. This allows the use of techniques originally devel-oped for MAP-MRF inference and learning. We explore two ways to set up the objective function - by discretizing the Bethe free energy and by learning it from training data. Experimental results show that for certain types of graphs a learned function can out-perform the Bethe approximation. We also establish a link between the Bethe free energy and submodular functions.","lang":"eng"}],"alternative_title":["IST Austria Technical Report"],"type":"technical_report","date_created":"2018-12-12T11:39:06Z","date_updated":"2023-02-23T11:13:22Z","file":[{"file_name":"IST-2012-0003_IST-2012-0003.pdf","access_level":"open_access","creator":"system","content_type":"application/pdf","file_size":618744,"file_id":"5490","relation":"main_file","date_updated":"2020-07-14T12:46:44Z","date_created":"2018-12-12T11:53:29Z","checksum":"7e0ba85ad123b13223aaf6cdde2d288c"}],"oa_version":"Published Version","author":[{"full_name":"Korc, Filip","last_name":"Korc","first_name":"Filip","id":"476A2FD6-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Kolmogorov, Vladimir","first_name":"Vladimir","last_name":"Kolmogorov","id":"3D50B0BA-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","full_name":"Lampert, Christoph"}],"pubrep_id":"36","related_material":{"record":[{"id":"3124","relation":"earlier_version","status":"public"}]},"ddc":["000"],"publication_status":"published","title":"Approximating marginals using discrete energy minimization","status":"public","department":[{"_id":"VlKo"},{"_id":"ChLa"}],"publisher":"IST Austria","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"5396","year":"2012","month":"07","day":"23","publication_identifier":{"issn":["2664-1690"]},"has_accepted_license":"1","language":[{"iso":"eng"}],"date_published":"2012-07-23T00:00:00Z","doi":"10.15479/AT:IST-2012-0003","page":"13","citation":{"chicago":"Korc, Filip, Vladimir Kolmogorov, and Christoph Lampert. Approximating Marginals Using Discrete Energy Minimization. IST Austria, 2012. https://doi.org/10.15479/AT:IST-2012-0003.","short":"F. Korc, V. Kolmogorov, C. Lampert, Approximating Marginals Using Discrete Energy Minimization, IST Austria, 2012.","mla":"Korc, Filip, et al. Approximating Marginals Using Discrete Energy Minimization. IST Austria, 2012, doi:10.15479/AT:IST-2012-0003.","ieee":"F. Korc, V. Kolmogorov, and C. Lampert, Approximating marginals using discrete energy minimization. IST Austria, 2012.","apa":"Korc, F., Kolmogorov, V., & Lampert, C. (2012). Approximating marginals using discrete energy minimization. IST Austria. https://doi.org/10.15479/AT:IST-2012-0003","ista":"Korc F, Kolmogorov V, Lampert C. 2012. Approximating marginals using discrete energy minimization, IST Austria, 13p.","ama":"Korc F, Kolmogorov V, Lampert C. Approximating Marginals Using Discrete Energy Minimization. IST Austria; 2012. doi:10.15479/AT:IST-2012-0003"},"oa":1},{"department":[{"_id":"ChLa"}],"publisher":"Deutsches Zentrum für Luft und Raumfahrt","status":"public","title":"Multi-modal learning for dynamic tactile sensing","publication_status":"published","acknowledgement":"The project receives funding from the European Community’s Seventh Framework Programme under grant agreement\r\nno. ICT- 248273 GeRT.","_id":"2915","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2012","oa_version":"None","date_created":"2018-12-11T12:00:19Z","date_updated":"2023-10-17T07:58:59Z","author":[{"first_name":"Oliver","last_name":"Kroemer","full_name":"Kroemer, Oliver"},{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"},{"full_name":"Peters, Jan","last_name":"Peters","first_name":"Jan"}],"type":"conference","publist_id":"3828","quality_controlled":"1","citation":{"chicago":"Kroemer, Oliver, Christoph Lampert, and Jan Peters. “Multi-Modal Learning for Dynamic Tactile Sensing.” Deutsches Zentrum für Luft und Raumfahrt, 2012.","short":"O. Kroemer, C. Lampert, J. Peters, in:, Deutsches Zentrum für Luft und Raumfahrt, 2012.","mla":"Kroemer, Oliver, et al. Multi-Modal Learning for Dynamic Tactile Sensing. Deutsches Zentrum für Luft und Raumfahrt, 2012.","ieee":"O. Kroemer, C. Lampert, and J. Peters, “Multi-modal learning for dynamic tactile sensing,” 2012.","apa":"Kroemer, O., Lampert, C., & Peters, J. (2012). Multi-modal learning for dynamic tactile sensing. Deutsches Zentrum für Luft und Raumfahrt.","ista":"Kroemer O, Lampert C, Peters J. 2012. Multi-modal learning for dynamic tactile sensing","ama":"Kroemer O, Lampert C, Peters J. Multi-modal learning for dynamic tactile sensing. In: Deutsches Zentrum für Luft und Raumfahrt; 2012."},"language":[{"iso":"eng"}],"date_published":"2012-10-11T00:00:00Z","article_processing_charge":"No","day":"11","month":"10"},{"author":[{"last_name":"Quadrianto","first_name":"Novi","full_name":"Quadrianto, Novi"},{"full_name":"Lampert, Christoph","last_name":"Lampert","first_name":"Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Chen, Chao","id":"3E92416E-F248-11E8-B48F-1D18A9856A87","last_name":"Chen","first_name":"Chao"}],"date_updated":"2023-10-17T11:55:06Z","date_created":"2018-12-11T12:01:33Z","oa_version":"Preprint","_id":"3127","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2012","publication_status":"published","title":"The most persistent soft-clique in a set of sampled graphs","status":"public","publisher":"ML Research Press","department":[{"_id":"ChLa"},{"_id":"HeEd"}],"abstract":[{"lang":"eng","text":"When searching for characteristic subpatterns in potentially noisy graph data, it appears self-evident that having multiple observations would be better than having just one. However, it turns out that the inconsistencies introduced when different graph instances have different edge sets pose a serious challenge. In this work we address this challenge for the problem of finding maximum weighted cliques.\r\n We introduce the concept of most persistent soft-clique. This is subset of vertices, that 1) is almost fully or at least densely connected, 2) occurs in all or almost all graph instances, and 3) has the maximum weight. We present a measure of clique-ness, that essentially counts the number of edge missing to make a subset of vertices into a clique. With this measure, we show that the problem of finding the most persistent soft-clique problem can be cast either as: a) a max-min two person game optimization problem, or b) a min-min soft margin optimization problem. Both formulations lead to the same solution when using a partial Lagrangian method to solve the optimization problems. By experiments on synthetic data and on real social network data, we show that the proposed method is able to reliably find soft cliques in graph data, even if that is distorted by random noise or unreliable observations."}],"publist_id":"3572","type":"conference","conference":{"end_date":"2012-07-01","location":"Edinburgh, United Kingdom","start_date":"2012-06-26","name":"ICML: International Conference on Machine Learning"},"date_published":"2012-06-01T00:00:00Z","language":[{"iso":"eng"}],"publication":"Proceedings of the 29th International Conference on Machine Learning","citation":{"mla":"Quadrianto, Novi, et al. “The Most Persistent Soft-Clique in a Set of Sampled Graphs.” Proceedings of the 29th International Conference on Machine Learning, ML Research Press, 2012, pp. 211–18.","short":"N. Quadrianto, C. Lampert, C. Chen, in:, Proceedings of the 29th International Conference on Machine Learning, ML Research Press, 2012, pp. 211–218.","chicago":"Quadrianto, Novi, Christoph Lampert, and Chao Chen. “The Most Persistent Soft-Clique in a Set of Sampled Graphs.” In Proceedings of the 29th International Conference on Machine Learning, 211–18. ML Research Press, 2012.","ama":"Quadrianto N, Lampert C, Chen C. The most persistent soft-clique in a set of sampled graphs. In: Proceedings of the 29th International Conference on Machine Learning. ML Research Press; 2012:211-218.","ista":"Quadrianto N, Lampert C, Chen C. 2012. The most persistent soft-clique in a set of sampled graphs. Proceedings of the 29th International Conference on Machine Learning. ICML: International Conference on Machine Learning, 211–218.","ieee":"N. Quadrianto, C. Lampert, and C. Chen, “The most persistent soft-clique in a set of sampled graphs,” in Proceedings of the 29th International Conference on Machine Learning, Edinburgh, United Kingdom, 2012, pp. 211–218.","apa":"Quadrianto, N., Lampert, C., & Chen, C. (2012). The most persistent soft-clique in a set of sampled graphs. In Proceedings of the 29th International Conference on Machine Learning (pp. 211–218). Edinburgh, United Kingdom: ML Research Press."},"main_file_link":[{"url":"http://arxiv.org/abs/1206.4652","open_access":"1"}],"oa":1,"quality_controlled":"1","page":"211-218","day":"01","month":"06","article_processing_charge":"No","scopus_import":"1"},{"abstract":[{"text":"Playing table tennis is a difficult task for robots, especially due to their limitations of acceleration. A key bottleneck is the amount of time needed to reach the desired hitting position and velocity of the racket for returning the incoming ball. Here, it often does not suffice to simply extrapolate the ball's trajectory after the opponent returns it but more information is needed. Humans are able to predict the ball's trajectory based on the opponent's moves and, thus, have a considerable advantage. Hence, we propose to incorporate an anticipation system into robot table tennis players, which enables the robot to react earlier while the opponent is performing the striking movement. Based on visual observation of the opponent's racket movement, the robot can predict the aim of the opponent and adjust its movement generation accordingly. The policies for deciding how and when to react are obtained by reinforcement learning. We conduct experiments with an existing robot player to show that the learned reaction policy can significantly improve the performance of the overall system.","lang":"eng"}],"publist_id":"3293","type":"conference","date_created":"2018-12-11T12:02:45Z","date_updated":"2021-01-12T07:42:45Z","oa_version":"None","author":[{"full_name":"Wang, Zhikun","first_name":"Zhikun","last_name":"Wang"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"},{"full_name":"Mülling, Katharina","last_name":"Mülling","first_name":"Katharina"},{"first_name":"Bernhard","last_name":"Schölkopf","full_name":"Schölkopf, Bernhard"},{"full_name":"Peters, Jan","last_name":"Peters","first_name":"Jan"}],"status":"public","publication_status":"published","title":"Learning anticipation policies for robot table tennis","department":[{"_id":"ChLa"}],"publisher":"IEEE","year":"2011","_id":"3337","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","month":"01","day":"01","scopus_import":1,"language":[{"iso":"eng"}],"conference":{"start_date":"2011-09-25","location":"San Francisco, USA","end_date":"2011-09-30","name":"IROS: RSJ International Conference on Intelligent Robots and Systems"},"date_published":"2011-01-01T00:00:00Z","doi":"10.1109/IROS.2011.6094892","quality_controlled":"1","page":"332 - 337","citation":{"chicago":"Wang, Zhikun, Christoph Lampert, Katharina Mülling, Bernhard Schölkopf, and Jan Peters. “Learning Anticipation Policies for Robot Table Tennis,” 332–37. IEEE, 2011. https://doi.org/10.1109/IROS.2011.6094892.","short":"Z. Wang, C. Lampert, K. Mülling, B. Schölkopf, J. Peters, in:, IEEE, 2011, pp. 332–337.","mla":"Wang, Zhikun, et al. Learning Anticipation Policies for Robot Table Tennis. IEEE, 2011, pp. 332–37, doi:10.1109/IROS.2011.6094892.","apa":"Wang, Z., Lampert, C., Mülling, K., Schölkopf, B., & Peters, J. (2011). Learning anticipation policies for robot table tennis (pp. 332–337). Presented at the IROS: RSJ International Conference on Intelligent Robots and Systems, San Francisco, USA: IEEE. https://doi.org/10.1109/IROS.2011.6094892","ieee":"Z. Wang, C. Lampert, K. Mülling, B. Schölkopf, and J. Peters, “Learning anticipation policies for robot table tennis,” presented at the IROS: RSJ International Conference on Intelligent Robots and Systems, San Francisco, USA, 2011, pp. 332–337.","ista":"Wang Z, Lampert C, Mülling K, Schölkopf B, Peters J. 2011. Learning anticipation policies for robot table tennis. IROS: RSJ International Conference on Intelligent Robots and Systems, 332–337.","ama":"Wang Z, Lampert C, Mülling K, Schölkopf B, Peters J. Learning anticipation policies for robot table tennis. In: IEEE; 2011:332-337. doi:10.1109/IROS.2011.6094892"}},{"doi":"10.1016/j.patrec.2011.02.011","date_published":"2011-08-01T00:00:00Z","language":[{"iso":"eng"}],"citation":{"ama":"Blaschko M, Shelton J, Bartels A, Lampert C, Gretton A. Semi supervised kernel canonical correlation analysis with application to human fMRI. Pattern Recognition Letters. 2011;32(11):1572-1583. doi:10.1016/j.patrec.2011.02.011","apa":"Blaschko, M., Shelton, J., Bartels, A., Lampert, C., & Gretton, A. (2011). Semi supervised kernel canonical correlation analysis with application to human fMRI. Pattern Recognition Letters. Elsevier. https://doi.org/10.1016/j.patrec.2011.02.011","ieee":"M. Blaschko, J. Shelton, A. Bartels, C. Lampert, and A. Gretton, “Semi supervised kernel canonical correlation analysis with application to human fMRI,” Pattern Recognition Letters, vol. 32, no. 11. Elsevier, pp. 1572–1583, 2011.","ista":"Blaschko M, Shelton J, Bartels A, Lampert C, Gretton A. 2011. Semi supervised kernel canonical correlation analysis with application to human fMRI. Pattern Recognition Letters. 32(11), 1572–1583.","short":"M. Blaschko, J. Shelton, A. Bartels, C. Lampert, A. Gretton, Pattern Recognition Letters 32 (2011) 1572–1583.","mla":"Blaschko, Matthew, et al. “Semi Supervised Kernel Canonical Correlation Analysis with Application to Human FMRI.” Pattern Recognition Letters, vol. 32, no. 11, Elsevier, 2011, pp. 1572–83, doi:10.1016/j.patrec.2011.02.011.","chicago":"Blaschko, Matthew, Jacquelyn Shelton, Andreas Bartels, Christoph Lampert, and Arthur Gretton. “Semi Supervised Kernel Canonical Correlation Analysis with Application to Human FMRI.” Pattern Recognition Letters. Elsevier, 2011. https://doi.org/10.1016/j.patrec.2011.02.011."},"publication":"Pattern Recognition Letters","page":"1572 - 1583","quality_controlled":"1","month":"08","day":"01","scopus_import":1,"author":[{"full_name":"Blaschko, Matthew","first_name":"Matthew","last_name":"Blaschko"},{"first_name":"Jacquelyn","last_name":"Shelton","full_name":"Shelton, Jacquelyn"},{"first_name":"Andreas","last_name":"Bartels","full_name":"Bartels, Andreas"},{"first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","full_name":"Lampert, Christoph"},{"full_name":"Gretton, Arthur","first_name":"Arthur","last_name":"Gretton"}],"oa_version":"None","volume":32,"date_created":"2018-12-11T12:03:03Z","date_updated":"2021-01-12T07:43:09Z","_id":"3389","acknowledgement":"The research leading to these results has received funding from the European Research Council under the European Community’s Seventh Framework Programme (FP7/2007-2013)/ERC Grant Agreement No. 228180. This work was funded in part by the EC project CLASS, IST 027978, and the PASCAL2 network of excellence, IST 2002-506778.","year":"2011","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","intvolume":" 32","department":[{"_id":"ChLa"}],"publisher":"Elsevier","status":"public","title":"Semi supervised kernel canonical correlation analysis with application to human fMRI","publication_status":"published","publist_id":"3218","issue":"11","abstract":[{"lang":"eng","text":"Kernel canonical correlation analysis (KCCA) is a general technique for subspace learning that incorporates principal components analysis (PCA) and Fisher linear discriminant analysis (LDA) as special cases. By finding directions that maximize correlation, KCCA learns representations that are more closely tied to the underlying process that generates the data and can ignore high-variance noise directions. However, for data where acquisition in one or more modalities is expensive or otherwise limited, KCCA may suffer from small sample effects. We propose to use semi-supervised Laplacian regularization to utilize data that are present in only one modality. This approach is able to find highly correlated directions that also lie along the data manifold, resulting in a more robust estimate of correlated subspaces. Functional magnetic resonance imaging (fMRI) acquired data are naturally amenable to subspace techniques as data are well aligned. fMRI data of the human brain are a particularly interesting candidate. In this study we implemented various supervised and semi-supervised versions of KCCA on human fMRI data, with regression to single and multi-variate labels (corresponding to video content subjects viewed during the image acquisition). In each variate condition, the semi-supervised variants of KCCA performed better than the supervised variants, including a supervised variant with Laplacian regularization. We additionally analyze the weights learned by the regression in order to infer brain regions that are important to different types of visual processing."}],"type":"journal_article"},{"abstract":[{"lang":"eng","text":"Dynamic tactile sensing is a fundamental ability to recognize materials and objects. However, while humans are born with partially developed dynamic tactile sensing and quickly master this skill, today's robots remain in their infancy. The development of such a sense requires not only better sensors but the right algorithms to deal with these sensors' data as well. For example, when classifying a material based on touch, the data are noisy, high-dimensional, and contain irrelevant signals as well as essential ones. Few classification methods from machine learning can deal with such problems. In this paper, we propose an efficient approach to infer suitable lower dimensional representations of the tactile data. In order to classify materials based on only the sense of touch, these representations are autonomously discovered using visual information of the surfaces during training. However, accurately pairing vision and tactile samples in real-robot applications is a difficult problem. The proposed approach, therefore, works with weak pairings between the modalities. Experiments show that the resulting approach is very robust and yields significantly higher classification performance based on only dynamic tactile sensing."}],"publist_id":"3225","issue":"3","type":"journal_article","author":[{"full_name":"Kroemer, Oliver","last_name":"Kroemer","first_name":"Oliver"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert","full_name":"Lampert, Christoph"},{"full_name":"Peters, Jan","last_name":"Peters","first_name":"Jan"}],"date_created":"2018-12-11T12:03:01Z","date_updated":"2021-01-12T07:43:06Z","volume":27,"oa_version":"None","_id":"3382","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","year":"2011","publication_status":"published","title":"Learning dynamic tactile sensing with robust vision based training","status":"public","intvolume":" 27","department":[{"_id":"ChLa"}],"publisher":"IEEE","day":"21","month":"05","scopus_import":1,"date_published":"2011-05-21T00:00:00Z","doi":"10.1109/TRO.2011.2121130","language":[{"iso":"eng"}],"publication":"IEEE Transactions on Robotics","citation":{"chicago":"Kroemer, Oliver, Christoph Lampert, and Jan Peters. “Learning Dynamic Tactile Sensing with Robust Vision Based Training.” IEEE Transactions on Robotics. IEEE, 2011. https://doi.org/10.1109/TRO.2011.2121130.","short":"O. Kroemer, C. Lampert, J. Peters, IEEE Transactions on Robotics 27 (2011) 545–557.","mla":"Kroemer, Oliver, et al. “Learning Dynamic Tactile Sensing with Robust Vision Based Training.” IEEE Transactions on Robotics, vol. 27, no. 3, IEEE, 2011, pp. 545–57, doi:10.1109/TRO.2011.2121130.","apa":"Kroemer, O., Lampert, C., & Peters, J. (2011). Learning dynamic tactile sensing with robust vision based training. IEEE Transactions on Robotics. IEEE. https://doi.org/10.1109/TRO.2011.2121130","ieee":"O. Kroemer, C. Lampert, and J. Peters, “Learning dynamic tactile sensing with robust vision based training,” IEEE Transactions on Robotics, vol. 27, no. 3. IEEE, pp. 545–557, 2011.","ista":"Kroemer O, Lampert C, Peters J. 2011. Learning dynamic tactile sensing with robust vision based training. IEEE Transactions on Robotics. 27(3), 545–557.","ama":"Kroemer O, Lampert C, Peters J. Learning dynamic tactile sensing with robust vision based training. IEEE Transactions on Robotics. 2011;27(3):545-557. doi:10.1109/TRO.2011.2121130"},"quality_controlled":"1","page":"545 - 557"},{"citation":{"ista":"Chen C, Freedman D, Lampert C. 2011. Enforcing topological constraints in random field image segmentation, IST Austria, 69p.","ieee":"C. Chen, D. Freedman, and C. Lampert, Enforcing topological constraints in random field image segmentation. IST Austria, 2011.","apa":"Chen, C., Freedman, D., & Lampert, C. (2011). Enforcing topological constraints in random field image segmentation. IST Austria. https://doi.org/10.15479/AT:IST-2011-0002","ama":"Chen C, Freedman D, Lampert C. Enforcing Topological Constraints in Random Field Image Segmentation. IST Austria; 2011. doi:10.15479/AT:IST-2011-0002","chicago":"Chen, Chao, Daniel Freedman, and Christoph Lampert. Enforcing Topological Constraints in Random Field Image Segmentation. IST Austria, 2011. https://doi.org/10.15479/AT:IST-2011-0002.","mla":"Chen, Chao, et al. Enforcing Topological Constraints in Random Field Image Segmentation. IST Austria, 2011, doi:10.15479/AT:IST-2011-0002.","short":"C. Chen, D. Freedman, C. Lampert, Enforcing Topological Constraints in Random Field Image Segmentation, IST Austria, 2011."},"oa":1,"page":"69","doi":"10.15479/AT:IST-2011-0002","date_published":"2011-03-28T00:00:00Z","language":[{"iso":"eng"}],"day":"28","month":"03","publication_identifier":{"issn":["2664-1690"]},"has_accepted_license":"1","year":"2011","_id":"5386","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","status":"public","ddc":["000"],"publication_status":"published","title":"Enforcing topological constraints in random field image segmentation","publisher":"IST Austria","department":[{"_id":"ChLa"}],"author":[{"full_name":"Chen, Chao","id":"3E92416E-F248-11E8-B48F-1D18A9856A87","last_name":"Chen","first_name":"Chao"},{"full_name":"Freedman, Daniel","last_name":"Freedman","first_name":"Daniel"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"pubrep_id":"22","related_material":{"record":[{"relation":"later_version","status":"public","id":"3336"}]},"date_updated":"2023-02-23T11:22:48Z","date_created":"2018-12-12T11:39:02Z","oa_version":"Published Version","file":[{"file_name":"IST-2011-0002_IST-2011-0002.pdf","access_level":"open_access","content_type":"application/pdf","file_size":26390601,"creator":"system","relation":"main_file","file_id":"5495","date_updated":"2020-07-14T12:46:41Z","date_created":"2018-12-12T11:53:34Z","checksum":"ad64c2add5fe2ad10e9d5c669f3f9526"}],"type":"technical_report","alternative_title":["IST Austria Technical Report"],"file_date_updated":"2020-07-14T12:46:41Z","abstract":[{"text":"We introduce TopoCut: a new way to integrate knowledge about topological properties (TPs) into random field image segmentation model. Instead of including TPs as additional constraints during minimization of the energy function, we devise an efficient algorithm for modifying the unary potentials such that the resulting segmentation is guaranteed with the desired properties. Our method is more flexible in the sense that it handles more topology constraints than previous methods, which were only able to enforce pairwise or global connectivity. In particular, our method is very fast, making it for the first time possible to enforce global topological properties in practical image segmentation tasks.","lang":"eng"}]},{"publication_status":"published","status":"public","title":"Enforcing topological constraints in random field image segmentation","department":[{"_id":"HeEd"},{"_id":"ChLa"}],"publisher":"IEEE","_id":"3336","acknowledgement":"The first author is supported by the Austrian Science Fund (FWF) grant No. P20134-N13. The authors would like to thank Sebastian Nowozin for helpful discussions.","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","year":"2011","date_updated":"2023-02-23T12:23:56Z","date_created":"2018-12-11T12:02:45Z","oa_version":"None","author":[{"full_name":"Chen, Chao","first_name":"Chao","last_name":"Chen","id":"3E92416E-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Freedman","first_name":"Daniel","full_name":"Freedman, Daniel"},{"full_name":"Lampert, Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887","first_name":"Christoph","last_name":"Lampert"}],"related_material":{"record":[{"id":"5386","status":"public","relation":"earlier_version"}]},"type":"conference","abstract":[{"lang":"eng","text":"We introduce TopoCut: a new way to integrate knowledge about topological properties (TPs) into random field image segmentation model. Instead of including TPs as additional constraints during minimization of the energy function, we devise an efficient algorithm for modifying the unary potentials such that the resulting segmentation is guaranteed with the desired properties. Our method is more flexible in the sense that it handles more topology constraints than previous methods, which were only able to enforce pairwise or global connectivity. In particular, our method is very fast, making it for the first time possible to enforce global topological properties in practical image segmentation tasks."}],"publist_id":"3294","quality_controlled":"1","page":"2089 - 2096","publication":"CVPR: Computer Vision and Pattern Recognition","citation":{"ama":"Chen C, Freedman D, Lampert C. Enforcing topological constraints in random field image segmentation. In: CVPR: Computer Vision and Pattern Recognition. IEEE; 2011:2089-2096. doi:10.1109/CVPR.2011.5995503","ista":"Chen C, Freedman D, Lampert C. 2011. Enforcing topological constraints in random field image segmentation. CVPR: Computer Vision and Pattern Recognition. CVPR: Conference on Computer Vision and Pattern Recognition, 2089–2096.","ieee":"C. Chen, D. Freedman, and C. Lampert, “Enforcing topological constraints in random field image segmentation,” in CVPR: Computer Vision and Pattern Recognition, Colorado Springs, CO, United States, 2011, pp. 2089–2096.","apa":"Chen, C., Freedman, D., & Lampert, C. (2011). Enforcing topological constraints in random field image segmentation. In CVPR: Computer Vision and Pattern Recognition (pp. 2089–2096). Colorado Springs, CO, United States: IEEE. https://doi.org/10.1109/CVPR.2011.5995503","mla":"Chen, Chao, et al. “Enforcing Topological Constraints in Random Field Image Segmentation.” CVPR: Computer Vision and Pattern Recognition, IEEE, 2011, pp. 2089–96, doi:10.1109/CVPR.2011.5995503.","short":"C. Chen, D. Freedman, C. Lampert, in:, CVPR: Computer Vision and Pattern Recognition, IEEE, 2011, pp. 2089–2096.","chicago":"Chen, Chao, Daniel Freedman, and Christoph Lampert. “Enforcing Topological Constraints in Random Field Image Segmentation.” In CVPR: Computer Vision and Pattern Recognition, 2089–96. IEEE, 2011. https://doi.org/10.1109/CVPR.2011.5995503."},"language":[{"iso":"eng"}],"conference":{"end_date":"2011-06-25","location":"Colorado Springs, CO, United States","start_date":"2011-06-20","name":"CVPR: Conference on Computer Vision and Pattern Recognition"},"date_published":"2011-07-22T00:00:00Z","doi":"10.1109/CVPR.2011.5995503","scopus_import":"1","day":"22","month":"07","publication_identifier":{"eisbn":["978-1-4577-0395-9"],"isbn":["978-1-4577-0394-2"]},"article_processing_charge":"No"},{"scopus_import":1,"day":"01","month":"12","quality_controlled":"1","citation":{"short":"C. Lampert, in:, Neural Information Processing Systems, 2011.","mla":"Lampert, Christoph. Maximum Margin Multi-Label Structured Prediction. Neural Information Processing Systems, 2011.","chicago":"Lampert, Christoph. “Maximum Margin Multi-Label Structured Prediction.” Neural Information Processing Systems, 2011.","ama":"Lampert C. Maximum margin multi-label structured prediction. In: Neural Information Processing Systems; 2011.","ieee":"C. Lampert, “Maximum margin multi-label structured prediction,” presented at the NIPS: Neural Information Processing Systems, Granada, Spain, 2011.","apa":"Lampert, C. (2011). Maximum margin multi-label structured prediction. Presented at the NIPS: Neural Information Processing Systems, Granada, Spain: Neural Information Processing Systems.","ista":"Lampert C. 2011. Maximum margin multi-label structured prediction. NIPS: Neural Information Processing Systems."},"language":[{"iso":"eng"}],"date_published":"2011-12-01T00:00:00Z","conference":{"location":"Granada, Spain","start_date":"2011-12-12","end_date":"2011-12-14","name":"NIPS: Neural Information Processing Systems"},"type":"conference","publist_id":"3522","abstract":[{"text":"We study multi-label prediction for structured output sets, a problem that occurs, for example, in object detection in images, secondary structure prediction in computational biology, and graph matching with symmetries. Conventional multilabel classification techniques are typically not applicable in this situation, because they require explicit enumeration of the label set, which is infeasible in case of structured outputs. Relying on techniques originally designed for single-label structured prediction, in particular structured support vector machines, results in reduced prediction accuracy, or leads to infeasible optimization problems. In this work we derive a maximum-margin training formulation for multi-label structured prediction that remains computationally tractable while achieving high prediction accuracy. It also shares most beneficial properties with single-label maximum-margin approaches, in particular formulation as a convex optimization problem, efficient working set training, and PAC-Bayesian generalization bounds.","lang":"eng"}],"department":[{"_id":"ChLa"}],"publisher":"Neural Information Processing Systems","title":"Maximum margin multi-label structured prediction","status":"public","publication_status":"published","_id":"3163","year":"2011","user_id":"4435EBFC-F248-11E8-B48F-1D18A9856A87","oa_version":"None","date_created":"2018-12-11T12:01:45Z","date_updated":"2023-10-17T11:47:35Z","related_material":{"record":[{"relation":"later_version","status":"public","id":"3322"}]},"author":[{"full_name":"Lampert, Christoph","first_name":"Christoph","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-8622-7887"}]},{"publication":"NIPS: Neural Information Processing Systems","citation":{"ama":"Lampert C. Maximum Margin Multi Label Structured Prediction. Neural Information Processing Systems Foundation; 2011.","apa":"Lampert, C. (2011). Maximum margin multi label structured prediction. NIPS: Neural Information Processing Systems. Neural Information Processing Systems Foundation.","ieee":"C. Lampert, Maximum margin multi label structured prediction. Neural Information Processing Systems Foundation, 2011.","ista":"Lampert C. 2011. Maximum margin multi label structured prediction, Neural Information Processing Systems Foundation,p.","short":"C. Lampert, Maximum Margin Multi Label Structured Prediction, Neural Information Processing Systems Foundation, 2011.","mla":"Lampert, Christoph. “Maximum Margin Multi Label Structured Prediction.” NIPS: Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2011.","chicago":"Lampert, Christoph. Maximum Margin Multi Label Structured Prediction. NIPS: Neural Information Processing Systems. Neural Information Processing Systems Foundation, 2011."},"date_published":"2011-12-13T00:00:00Z","language":[{"iso":"eng"}],"day":"13","month":"12","article_processing_charge":"No","_id":"3322","year":"2011","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","status":"public","publication_status":"published","title":"Maximum margin multi label structured prediction","publisher":"Neural Information Processing Systems Foundation","department":[{"_id":"ChLa"}],"author":[{"full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph"}],"related_material":{"record":[{"status":"public","relation":"earlier_version","id":"3163"}]},"date_updated":"2023-10-17T11:47:36Z","date_created":"2018-12-11T12:02:40Z","oa_version":"None","type":"conference_poster","abstract":[{"text":"We study multi-label prediction for structured output spaces, a problem that occurs, for example, in object detection in images, secondary structure prediction in computational biology, and graph matching with symmetries. Conventional multi-label classification techniques are typically not applicable in this situation, because they require explicit enumeration of the label space, which is infeasible in case of structured outputs. Relying on techniques originally designed for single- label structured prediction, in particular structured support vector machines, results in reduced prediction accuracy, or leads to infeasible optimization problems. In this work we derive a maximum-margin training formulation for multi-label structured prediction that remains computationally tractable while achieving high prediction accuracy. It also shares most beneficial properties with single-label maximum-margin approaches, in particular a formulation as a convex optimization problem, efficient working set training, and PAC-Bayesian generalization bounds.","lang":"eng"}],"publist_id":"3313"}]