[{"month":"05","alternative_title":["ISTA Thesis"],"oa_version":"Published Version","abstract":[{"lang":"eng","text":"Deep learning has enabled breakthroughs in challenging computing problems and has emerged as the standard problem-solving tool for computer vision and natural language processing tasks.\r\nOne exception to this trend is safety-critical tasks where robustness and resilience requirements contradict the black-box nature of neural networks. \r\nTo deploy deep learning methods for these tasks, it is vital to provide guarantees on neural network agents' safety and robustness criteria. \r\nThis can be achieved by developing formal verification methods to verify the safety and robustness properties of neural networks.\r\n\r\nOur goal is to design, develop and assess safety verification methods for neural networks to improve their reliability and trustworthiness in real-world applications.\r\nThis thesis establishes techniques for the verification of compressed and adversarially trained models as well as the design of novel neural networks for verifiably safe decision-making.\r\n\r\nFirst, we establish the problem of verifying quantized neural networks. Quantization is a technique that trades numerical precision for the computational efficiency of running a neural network and is widely adopted in industry.\r\nWe show that neglecting the reduced precision when verifying a neural network can lead to wrong conclusions about the robustness and safety of the network, highlighting that novel techniques for quantized network verification are necessary. We introduce several bit-exact verification methods explicitly designed for quantized neural networks and experimentally confirm on realistic networks that the network's robustness and other formal properties are affected by the quantization.\r\n\r\nFurthermore, we perform a case study providing evidence that adversarial training, a standard technique for making neural networks more robust, has detrimental effects on the network's performance. This robustness-accuracy tradeoff has been studied before regarding the accuracy obtained on classification datasets where each data point is independent of all other data points. On the other hand, we investigate the tradeoff empirically in robot learning settings where a both, a high accuracy and a high robustness, are desirable.\r\nOur results suggest that the negative side-effects of adversarial training outweigh its robustness benefits in practice.\r\n\r\nFinally, we consider the problem of verifying safety when running a Bayesian neural network policy in a feedback loop with systems over the infinite time horizon. Bayesian neural networks are probabilistic models for learning uncertainties in the data and are therefore often used on robotic and healthcare applications where data is inherently stochastic.\r\nWe introduce a method for recalibrating Bayesian neural networks so that they yield probability distributions over safe decisions only.\r\nOur method learns a safety certificate that guarantees safety over the infinite time horizon to determine which decisions are safe in every possible state of the system.\r\nWe demonstrate the effectiveness of our approach on a series of reinforcement learning benchmarks."}],"ec_funded":1,"related_material":{"record":[{"relation":"part_of_dissertation","id":"10665","status":"public"},{"id":"10667","status":"public","relation":"part_of_dissertation"},{"relation":"part_of_dissertation","id":"11366","status":"public"},{"relation":"part_of_dissertation","status":"public","id":"7808"},{"relation":"part_of_dissertation","status":"public","id":"10666"}]},"language":[{"iso":"eng"}],"file":[{"creator":"mlechner","file_size":13210143,"date_updated":"2022-05-13T12:49:00Z","file_name":"src.zip","date_created":"2022-05-13T12:33:26Z","relation":"source_file","access_level":"closed","content_type":"application/zip","file_id":"11378","checksum":"8eefa9c7c10ca7e1a2ccdd731962a645"},{"checksum":"1b9e1e5a9a83ed9d89dad2f5133dc026","file_id":"11382","content_type":"application/pdf","access_level":"open_access","relation":"main_file","date_created":"2022-05-16T08:02:28Z","file_name":"thesis_main-a2.pdf","date_updated":"2022-05-17T15:19:39Z","file_size":2732536,"creator":"mlechner"}],"publication_status":"published","degree_awarded":"PhD","publication_identifier":{"isbn":["978-3-99078-017-6"]},"keyword":["neural networks","verification","machine learning"],"status":"public","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by-nd/4.0/legalcode","image":"/image/cc_by_nd.png","name":"Creative Commons Attribution-NoDerivatives 4.0 International (CC BY-ND 4.0)","short":"CC BY-ND (4.0)"},"type":"dissertation","_id":"11362","department":[{"_id":"GradSch"},{"_id":"ToHe"}],"file_date_updated":"2022-05-17T15:19:39Z","ddc":["004"],"date_updated":"2023-08-17T06:58:38Z","supervisor":[{"first_name":"Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724","last_name":"Henzinger"}],"oa":1,"publisher":"Institute of Science and Technology Austria","date_created":"2022-05-12T07:14:01Z","date_published":"2022-05-12T00:00:00Z","doi":"10.15479/at:ista:11362","page":"124","day":"12","year":"2022","has_accepted_license":"1","project":[{"name":"The Wittgenstein Prize","grant_number":"Z211","call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425"},{"name":"Vigilant Algorithmic Monitoring of Software","grant_number":"101020093","_id":"62781420-2b32-11ec-9570-8d9b63373d4d","call_identifier":"H2020"}],"title":"Learning verifiable representations","article_processing_charge":"No","author":[{"id":"3DC22916-F248-11E8-B48F-1D18A9856A87","first_name":"Mathias","last_name":"Lechner","full_name":"Lechner, Mathias"}],"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","citation":{"mla":"Lechner, Mathias. Learning Verifiable Representations. Institute of Science and Technology Austria, 2022, doi:10.15479/at:ista:11362.","apa":"Lechner, M. (2022). Learning verifiable representations. Institute of Science and Technology Austria. https://doi.org/10.15479/at:ista:11362","ama":"Lechner M. Learning verifiable representations. 2022. doi:10.15479/at:ista:11362","short":"M. Lechner, Learning Verifiable Representations, Institute of Science and Technology Austria, 2022.","ieee":"M. Lechner, “Learning verifiable representations,” Institute of Science and Technology Austria, 2022.","chicago":"Lechner, Mathias. “Learning Verifiable Representations.” Institute of Science and Technology Austria, 2022. https://doi.org/10.15479/at:ista:11362.","ista":"Lechner M. 2022. Learning verifiable representations. Institute of Science and Technology Austria."}},{"project":[{"_id":"62781420-2b32-11ec-9570-8d9b63373d4d","call_identifier":"H2020","grant_number":"101020093","name":"Vigilant Algorithmic Monitoring of Software"}],"external_id":{"isi":["000870310500006"],"arxiv":["2207.13549"]},"article_processing_charge":"No","author":[{"last_name":"Doveri","full_name":"Doveri, Kyveli","first_name":"Kyveli"},{"first_name":"Pierre","full_name":"Ganty, Pierre","last_name":"Ganty"},{"last_name":"Mazzocchi","full_name":"Mazzocchi, Nicolas Adrien","id":"b26baa86-3308-11ec-87b0-8990f34baa85","first_name":"Nicolas Adrien"}],"title":"FORQ-based language inclusion formal testing","citation":{"ista":"Doveri K, Ganty P, Mazzocchi NA. 2022. FORQ-based language inclusion formal testing. Computer Aided Verification. CAV: Computer Aided Verification, LNCS, vol. 13372, 109–129.","chicago":"Doveri, Kyveli, Pierre Ganty, and Nicolas Adrien Mazzocchi. “FORQ-Based Language Inclusion Formal Testing.” In Computer Aided Verification, 13372:109–29. Springer Nature, 2022. https://doi.org/10.1007/978-3-031-13188-2_6.","ieee":"K. Doveri, P. Ganty, and N. A. Mazzocchi, “FORQ-based language inclusion formal testing,” in Computer Aided Verification, Haifa, Israel, 2022, vol. 13372, pp. 109–129.","short":"K. Doveri, P. Ganty, N.A. Mazzocchi, in:, Computer Aided Verification, Springer Nature, 2022, pp. 109–129.","apa":"Doveri, K., Ganty, P., & Mazzocchi, N. A. (2022). FORQ-based language inclusion formal testing. In Computer Aided Verification (Vol. 13372, pp. 109–129). Haifa, Israel: Springer Nature. https://doi.org/10.1007/978-3-031-13188-2_6","ama":"Doveri K, Ganty P, Mazzocchi NA. FORQ-based language inclusion formal testing. In: Computer Aided Verification. Vol 13372. Springer Nature; 2022:109-129. doi:10.1007/978-3-031-13188-2_6","mla":"Doveri, Kyveli, et al. “FORQ-Based Language Inclusion Formal Testing.” Computer Aided Verification, vol. 13372, Springer Nature, 2022, pp. 109–29, doi:10.1007/978-3-031-13188-2_6."},"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","oa":1,"publisher":"Springer Nature","quality_controlled":"1","acknowledgement":"This work was partially funded by the ESF Investing in your future, the Madrid regional project S2018/TCS-4339 BLOQUES, the Spanish project PGC2018-102210-B-I00 BOSCO, the Ramón y Cajal fellowship RYC-2016-20281, and the ERC grant PR1001ERC02.","page":"109-129","date_created":"2023-01-16T10:06:31Z","doi":"10.1007/978-3-031-13188-2_6","date_published":"2022-08-06T00:00:00Z","year":"2022","isi":1,"has_accepted_license":"1","publication":"Computer Aided Verification","day":"06","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)"},"conference":{"start_date":"2022-08-07","location":"Haifa, Israel","end_date":"2022-08-10","name":"CAV: Computer Aided Verification"},"type":"conference","status":"public","_id":"12302","file_date_updated":"2023-01-30T12:51:02Z","department":[{"_id":"ToHe"}],"date_updated":"2023-09-05T15:13:36Z","ddc":["000"],"scopus_import":"1","alternative_title":["LNCS"],"intvolume":" 13372","month":"08","abstract":[{"lang":"eng","text":"We propose a novel algorithm to decide the language inclusion between (nondeterministic) Büchi automata, a PSPACE-complete problem. Our approach, like others before, leverage a notion of quasiorder to prune the search for a counterexample by discarding candidates which are subsumed by others for the quasiorder. Discarded candidates are guaranteed to not compromise the completeness of the algorithm. The novelty of our work lies in the quasiorder used to discard candidates. We introduce FORQs (family of right quasiorders) that we obtain by adapting the notion of family of right congruences put forward by Maler and Staiger in 1993. We define a FORQ-based inclusion algorithm which we prove correct and instantiate it for a specific FORQ, called the structural FORQ, induced by the Büchi automaton to the right of the inclusion sign. The resulting implementation, called FORKLIFT, scales up better than the state-of-the-art on a variety of benchmarks including benchmarks from program verification and theorem proving for word combinatorics. Artifact: https://doi.org/10.5281/zenodo.6552870"}],"oa_version":"Published Version","ec_funded":1,"volume":13372,"publication_status":"published","publication_identifier":{"eisbn":["9783031131882"],"isbn":["9783031131875"],"eissn":["1611-3349"],"issn":["0302-9743"]},"language":[{"iso":"eng"}],"file":[{"date_created":"2023-01-30T12:51:02Z","file_name":"2022_LNCS_Doveri.pdf","creator":"dernst","date_updated":"2023-01-30T12:51:02Z","file_size":497682,"file_id":"12465","checksum":"edc363b1be5447a09063e115c247918a","success":1,"access_level":"open_access","relation":"main_file","content_type":"application/pdf"}]},{"project":[{"call_identifier":"H2020","_id":"62781420-2b32-11ec-9570-8d9b63373d4d","grant_number":"101020093","name":"Vigilant Algorithmic Monitoring of Software"}],"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","citation":{"chicago":"Bose, Sougata, Thomas A Henzinger, Karoliina Lehtinen, Sven Schewe, and Patrick Totzke. “History-Deterministic Timed Automata Are Not Determinizable.” In 16th International Conference on Reachability Problems, 13608:67–76. Springer Nature, 2022. https://doi.org/10.1007/978-3-031-19135-0_5.","ista":"Bose S, Henzinger TA, Lehtinen K, Schewe S, Totzke P. 2022. History-deterministic timed automata are not determinizable. 16th International Conference on Reachability Problems. RC: Reachability Problems, LNCS, vol. 13608, 67–76.","mla":"Bose, Sougata, et al. “History-Deterministic Timed Automata Are Not Determinizable.” 16th International Conference on Reachability Problems, vol. 13608, Springer Nature, 2022, pp. 67–76, doi:10.1007/978-3-031-19135-0_5.","ieee":"S. Bose, T. A. Henzinger, K. Lehtinen, S. Schewe, and P. Totzke, “History-deterministic timed automata are not determinizable,” in 16th International Conference on Reachability Problems, Kaiserslautern, Germany, 2022, vol. 13608, pp. 67–76.","short":"S. Bose, T.A. Henzinger, K. Lehtinen, S. Schewe, P. Totzke, in:, 16th International Conference on Reachability Problems, Springer Nature, 2022, pp. 67–76.","ama":"Bose S, Henzinger TA, Lehtinen K, Schewe S, Totzke P. History-deterministic timed automata are not determinizable. In: 16th International Conference on Reachability Problems. Vol 13608. Springer Nature; 2022:67-76. doi:10.1007/978-3-031-19135-0_5","apa":"Bose, S., Henzinger, T. A., Lehtinen, K., Schewe, S., & Totzke, P. (2022). History-deterministic timed automata are not determinizable. In 16th International Conference on Reachability Problems (Vol. 13608, pp. 67–76). Kaiserslautern, Germany: Springer Nature. https://doi.org/10.1007/978-3-031-19135-0_5"},"title":"History-deterministic timed automata are not determinizable","article_processing_charge":"No","author":[{"last_name":"Bose","full_name":"Bose, Sougata","first_name":"Sougata"},{"first_name":"Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724","last_name":"Henzinger"},{"full_name":"Lehtinen, Karoliina","last_name":"Lehtinen","first_name":"Karoliina"},{"first_name":"Sven","full_name":"Schewe, Sven","last_name":"Schewe"},{"first_name":"Patrick","full_name":"Totzke, Patrick","last_name":"Totzke"}],"acknowledgement":"This work was supported in part by the ERC-2020-AdG 101020093, the EPSRC project EP/V025848/1, and the EPSRC project EP/X017796/1.","oa":1,"quality_controlled":"1","publisher":"Springer Nature","publication":"16th International Conference on Reachability Problems","day":"12","year":"2022","date_created":"2023-01-12T12:11:57Z","date_published":"2022-10-12T00:00:00Z","doi":"10.1007/978-3-031-19135-0_5","page":"67-76","_id":"12175","status":"public","conference":{"end_date":"2022-10-21","location":"Kaiserslautern, Germany","start_date":"2022-10-17","name":"RC: Reachability Problems"},"type":"conference","date_updated":"2023-09-05T15:12:08Z","department":[{"_id":"ToHe"}],"oa_version":"Preprint","abstract":[{"lang":"eng","text":"An automaton is history-deterministic (HD) if one can safely resolve its non-deterministic choices on the fly. In a recent paper, Henzinger, Lehtinen and Totzke studied this in the context of Timed Automata [9], where it was conjectured that the class of timed ω-languages recognised by HD-timed automata strictly extends that of deterministic ones. We provide a proof for this fact."}],"intvolume":" 13608","month":"10","main_file_link":[{"url":"https://hal.science/hal-03849398/","open_access":"1"}],"scopus_import":"1","alternative_title":["LNCS"],"language":[{"iso":"eng"}],"publication_status":"published","publication_identifier":{"eissn":["1611-3349"],"isbn":["9783031191343"],"issn":["0302-9743"],"eisbn":["9783031191350"]},"ec_funded":1,"volume":13608},{"scopus_import":"1","main_file_link":[{"url":"https://arxiv.org/abs/2107.08467","open_access":"1"}],"month":"06","intvolume":" 36","abstract":[{"text":"We introduce a new statistical verification algorithm that formally quantifies the behavioral robustness of any time-continuous process formulated as a continuous-depth model. Our algorithm solves a set of global optimization (Go) problems over a given time horizon to construct a tight enclosure (Tube) of the set of all process executions starting from a ball of initial states. We call our algorithm GoTube. Through its construction, GoTube ensures that the bounding tube is conservative up to a desired probability and up to a desired tightness.\r\n GoTube is implemented in JAX and optimized to scale to complex continuous-depth neural network models. Compared to advanced reachability analysis tools for time-continuous neural networks, GoTube does not accumulate overapproximation errors between time steps and avoids the infamous wrapping effect inherent in symbolic techniques. We show that GoTube substantially outperforms state-of-the-art verification tools in terms of the size of the initial ball, speed, time-horizon, task completion, and scalability on a large set of experiments.\r\n GoTube is stable and sets the state-of-the-art in terms of its ability to scale to time horizons well beyond what has been previously possible.","lang":"eng"}],"oa_version":"Preprint","volume":36,"issue":"6","ec_funded":1,"publication_identifier":{"issn":["2159-5399"],"eissn":["2374-3468"],"isbn":["978577358350"]},"publication_status":"published","language":[{"iso":"eng"}],"type":"journal_article","article_type":"original","status":"public","keyword":["General Medicine"],"_id":"12510","department":[{"_id":"ToHe"}],"date_updated":"2023-09-26T10:46:59Z","publisher":"Association for the Advancement of Artificial Intelligence","quality_controlled":"1","oa":1,"acknowledgement":"SG is funded by the Austrian Science Fund (FWF) project number W1255-N23. ML and TH are supported in part by FWF under grant Z211-N23 (Wittgenstein Award) and the ERC-2020-AdG 101020093. SS is supported by NSF awards DCL-2040599, CCF-1918225, and CPS-1446832. RH and DR are partially supported by Boeing. RG is partially supported by Horizon-2020 ECSEL Project grant No. 783163 (iDev40).","page":"6755-6764","doi":"10.1609/aaai.v36i6.20631","date_published":"2022-06-28T00:00:00Z","date_created":"2023-02-05T17:27:42Z","year":"2022","day":"28","publication":"Proceedings of the AAAI Conference on Artificial Intelligence","project":[{"call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425","grant_number":"Z211","name":"The Wittgenstein Prize"},{"grant_number":"101020093","name":"Vigilant Algorithmic Monitoring of Software","call_identifier":"H2020","_id":"62781420-2b32-11ec-9570-8d9b63373d4d"}],"author":[{"last_name":"Gruenbacher","full_name":"Gruenbacher, Sophie A.","first_name":"Sophie A."},{"last_name":"Lechner","full_name":"Lechner, Mathias","first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Hasani","full_name":"Hasani, Ramin","first_name":"Ramin"},{"full_name":"Rus, Daniela","last_name":"Rus","first_name":"Daniela"},{"first_name":"Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","last_name":"Henzinger","full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724"},{"first_name":"Scott A.","full_name":"Smolka, Scott A.","last_name":"Smolka"},{"first_name":"Radu","full_name":"Grosu, Radu","last_name":"Grosu"}],"external_id":{"arxiv":["2107.08467"]},"article_processing_charge":"No","title":"GoTube: Scalable statistical verification of continuous-depth models","citation":{"short":"S.A. Gruenbacher, M. Lechner, R. Hasani, D. Rus, T.A. Henzinger, S.A. Smolka, R. Grosu, Proceedings of the AAAI Conference on Artificial Intelligence 36 (2022) 6755–6764.","ieee":"S. A. Gruenbacher et al., “GoTube: Scalable statistical verification of continuous-depth models,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 6. Association for the Advancement of Artificial Intelligence, pp. 6755–6764, 2022.","ama":"Gruenbacher SA, Lechner M, Hasani R, et al. GoTube: Scalable statistical verification of continuous-depth models. Proceedings of the AAAI Conference on Artificial Intelligence. 2022;36(6):6755-6764. doi:10.1609/aaai.v36i6.20631","apa":"Gruenbacher, S. A., Lechner, M., Hasani, R., Rus, D., Henzinger, T. A., Smolka, S. A., & Grosu, R. (2022). GoTube: Scalable statistical verification of continuous-depth models. Proceedings of the AAAI Conference on Artificial Intelligence. Association for the Advancement of Artificial Intelligence. https://doi.org/10.1609/aaai.v36i6.20631","mla":"Gruenbacher, Sophie A., et al. “GoTube: Scalable Statistical Verification of Continuous-Depth Models.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 6, Association for the Advancement of Artificial Intelligence, 2022, pp. 6755–64, doi:10.1609/aaai.v36i6.20631.","ista":"Gruenbacher SA, Lechner M, Hasani R, Rus D, Henzinger TA, Smolka SA, Grosu R. 2022. GoTube: Scalable statistical verification of continuous-depth models. Proceedings of the AAAI Conference on Artificial Intelligence. 36(6), 6755–6764.","chicago":"Gruenbacher, Sophie A., Mathias Lechner, Ramin Hasani, Daniela Rus, Thomas A Henzinger, Scott A. Smolka, and Radu Grosu. “GoTube: Scalable Statistical Verification of Continuous-Depth Models.” Proceedings of the AAAI Conference on Artificial Intelligence. Association for the Advancement of Artificial Intelligence, 2022. https://doi.org/10.1609/aaai.v36i6.20631."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87"},{"abstract":[{"text":"We consider the problem of formally verifying almost-sure (a.s.) asymptotic stability in discrete-time nonlinear stochastic control systems. While verifying stability in deterministic control systems is extensively studied in the literature, verifying stability in stochastic control systems is an open problem. The few existing works on this topic either consider only specialized forms of stochasticity or make restrictive assumptions on the system, rendering them inapplicable to learning algorithms with neural network policies. \r\n In this work, we present an approach for general nonlinear stochastic control problems with two novel aspects: (a) instead of classical stochastic extensions of Lyapunov functions, we use ranking supermartingales (RSMs) to certify a.s. asymptotic stability, and (b) we present a method for learning neural network RSMs. \r\n We prove that our approach guarantees a.s. asymptotic stability of the system and\r\n provides the first method to obtain bounds on the stabilization time, which stochastic Lyapunov functions do not.\r\n Finally, we validate our approach experimentally on a set of nonlinear stochastic reinforcement learning environments with neural network policies.","lang":"eng"}],"oa_version":"Preprint","scopus_import":"1","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2112.09495"}],"month":"06","intvolume":" 36","publication_identifier":{"issn":["2159-5399"],"eissn":["2374-3468"],"isbn":["9781577358350"]},"publication_status":"published","language":[{"iso":"eng"}],"issue":"7","volume":36,"related_material":{"record":[{"relation":"dissertation_contains","status":"public","id":"14539"}]},"ec_funded":1,"_id":"12511","type":"journal_article","article_type":"original","status":"public","keyword":["General Medicine"],"date_updated":"2023-11-30T10:55:37Z","department":[{"_id":"ToHe"},{"_id":"KrCh"}],"acknowledgement":"This work was supported in part by the ERC-2020-AdG 101020093, ERC CoG 863818 (FoRM-SMArt) and the European Union’s Horizon 2020 research and innovation programme\r\nunder the Marie Skłodowska-Curie Grant Agreement No. 665385.","quality_controlled":"1","publisher":"Association for the Advancement of Artificial Intelligence","oa":1,"year":"2022","day":"28","publication":"Proceedings of the AAAI Conference on Artificial Intelligence","page":"7326-7336","doi":"10.1609/aaai.v36i7.20695","date_published":"2022-06-28T00:00:00Z","date_created":"2023-02-05T17:29:50Z","project":[{"_id":"62781420-2b32-11ec-9570-8d9b63373d4d","call_identifier":"H2020","grant_number":"101020093","name":"Vigilant Algorithmic Monitoring of Software"},{"_id":"0599E47C-7A3F-11EA-A408-12923DDC885E","call_identifier":"H2020","grant_number":"863818","name":"Formal Methods for Stochastic Models: Algorithms and Applications"},{"_id":"2564DBCA-B435-11E9-9278-68D0E5697425","call_identifier":"H2020","grant_number":"665385","name":"International IST Doctoral Program"}],"citation":{"chicago":"Lechner, Mathias, Dorde Zikelic, Krishnendu Chatterjee, and Thomas A Henzinger. “Stability Verification in Stochastic Control Systems via Neural Network Supermartingales.” Proceedings of the AAAI Conference on Artificial Intelligence. Association for the Advancement of Artificial Intelligence, 2022. https://doi.org/10.1609/aaai.v36i7.20695.","ista":"Lechner M, Zikelic D, Chatterjee K, Henzinger TA. 2022. Stability verification in stochastic control systems via neural network supermartingales. Proceedings of the AAAI Conference on Artificial Intelligence. 36(7), 7326–7336.","mla":"Lechner, Mathias, et al. “Stability Verification in Stochastic Control Systems via Neural Network Supermartingales.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 7, Association for the Advancement of Artificial Intelligence, 2022, pp. 7326–36, doi:10.1609/aaai.v36i7.20695.","short":"M. Lechner, D. Zikelic, K. Chatterjee, T.A. Henzinger, Proceedings of the AAAI Conference on Artificial Intelligence 36 (2022) 7326–7336.","ieee":"M. Lechner, D. Zikelic, K. Chatterjee, and T. A. Henzinger, “Stability verification in stochastic control systems via neural network supermartingales,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 7. Association for the Advancement of Artificial Intelligence, pp. 7326–7336, 2022.","ama":"Lechner M, Zikelic D, Chatterjee K, Henzinger TA. Stability verification in stochastic control systems via neural network supermartingales. Proceedings of the AAAI Conference on Artificial Intelligence. 2022;36(7):7326-7336. doi:10.1609/aaai.v36i7.20695","apa":"Lechner, M., Zikelic, D., Chatterjee, K., & Henzinger, T. A. (2022). Stability verification in stochastic control systems via neural network supermartingales. Proceedings of the AAAI Conference on Artificial Intelligence. Association for the Advancement of Artificial Intelligence. https://doi.org/10.1609/aaai.v36i7.20695"},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","author":[{"full_name":"Lechner, Mathias","last_name":"Lechner","id":"3DC22916-F248-11E8-B48F-1D18A9856A87","first_name":"Mathias"},{"id":"294AA7A6-F248-11E8-B48F-1D18A9856A87","first_name":"Dorde","last_name":"Zikelic","orcid":"0000-0002-4681-1699","full_name":"Zikelic, Dorde"},{"orcid":"0000-0002-4561-241X","full_name":"Chatterjee, Krishnendu","last_name":"Chatterjee","first_name":"Krishnendu","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Henzinger","orcid":"0000-0002-2985-7724","full_name":"Henzinger, Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","first_name":"Thomas A"}],"article_processing_charge":"No","external_id":{"arxiv":["2112.09495"]},"title":"Stability verification in stochastic control systems via neural network supermartingales"},{"_id":"14601","status":"public","project":[{"call_identifier":"H2020","_id":"62781420-2b32-11ec-9570-8d9b63373d4d","grant_number":"101020093","name":"Vigilant Algorithmic Monitoring of Software"},{"call_identifier":"H2020","_id":"0599E47C-7A3F-11EA-A408-12923DDC885E","grant_number":"863818","name":"Formal Methods for Stochastic Models: Algorithms and Applications"},{"grant_number":"665385","name":"International IST Doctoral Program","call_identifier":"H2020","_id":"2564DBCA-B435-11E9-9278-68D0E5697425"}],"type":"preprint","user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","citation":{"apa":"Zikelic, D., Lechner, M., Chatterjee, K., & Henzinger, T. A. (n.d.). Learning stabilizing policies in stochastic control systems. arXiv. https://doi.org/10.48550/arXiv.2205.11991","ama":"Zikelic D, Lechner M, Chatterjee K, Henzinger TA. Learning stabilizing policies in stochastic control systems. arXiv. doi:10.48550/arXiv.2205.11991","ieee":"D. Zikelic, M. Lechner, K. Chatterjee, and T. A. Henzinger, “Learning stabilizing policies in stochastic control systems,” arXiv. .","short":"D. Zikelic, M. Lechner, K. Chatterjee, T.A. Henzinger, ArXiv (n.d.).","mla":"Zikelic, Dorde, et al. “Learning Stabilizing Policies in Stochastic Control Systems.” ArXiv, doi:10.48550/arXiv.2205.11991.","ista":"Zikelic D, Lechner M, Chatterjee K, Henzinger TA. Learning stabilizing policies in stochastic control systems. arXiv, 10.48550/arXiv.2205.11991.","chicago":"Zikelic, Dorde, Mathias Lechner, Krishnendu Chatterjee, and Thomas A Henzinger. “Learning Stabilizing Policies in Stochastic Control Systems.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2205.11991."},"date_updated":"2023-11-30T10:55:37Z","title":"Learning stabilizing policies in stochastic control systems","department":[{"_id":"KrCh"},{"_id":"ToHe"}],"author":[{"first_name":"Dorde","id":"294AA7A6-F248-11E8-B48F-1D18A9856A87","last_name":"Zikelic","orcid":"0000-0002-4681-1699","full_name":"Zikelic, Dorde"},{"first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87","full_name":"Lechner, Mathias","last_name":"Lechner"},{"full_name":"Chatterjee, Krishnendu","orcid":"0000-0002-4561-241X","last_name":"Chatterjee","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87","first_name":"Krishnendu"},{"id":"40876CD8-F248-11E8-B48F-1D18A9856A87","first_name":"Thomas A","last_name":"Henzinger","orcid":"0000-0002-2985-7724","full_name":"Henzinger, Thomas A"}],"external_id":{"arxiv":["2205.11991"]},"article_processing_charge":"No","oa_version":"Preprint","abstract":[{"text":"In this work, we address the problem of learning provably stable neural\r\nnetwork policies for stochastic control systems. While recent work has\r\ndemonstrated the feasibility of certifying given policies using martingale\r\ntheory, the problem of how to learn such policies is little explored. Here, we\r\nstudy the effectiveness of jointly learning a policy together with a martingale\r\ncertificate that proves its stability using a single learning algorithm. We\r\nobserve that the joint optimization problem becomes easily stuck in local\r\nminima when starting from a randomly initialized policy. Our results suggest\r\nthat some form of pre-training of the policy is required for the joint\r\noptimization to repair and verify the policy successfully.","lang":"eng"}],"month":"05","oa":1,"main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2205.11991"}],"day":"24","language":[{"iso":"eng"}],"publication":"arXiv","publication_status":"submitted","year":"2022","doi":"10.48550/arXiv.2205.11991","date_published":"2022-05-24T00:00:00Z","related_material":{"record":[{"relation":"dissertation_contains","id":"14539","status":"public"}]},"ec_funded":1,"date_created":"2023-11-24T13:22:30Z"},{"month":"11","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/2210.05308"}],"oa":1,"oa_version":"Preprint","abstract":[{"lang":"eng","text":"We study the problem of learning controllers for discrete-time non-linear stochastic dynamical systems with formal reach-avoid guarantees. This work presents the first method for providing formal reach-avoid guarantees, which combine and generalize stability and safety guarantees, with a tolerable probability threshold $p\\in[0,1]$ over the infinite time horizon. Our method leverages advances in machine learning literature and it represents formal certificates as neural networks. In particular, we learn a certificate in the form of a reach-avoid supermartingale (RASM), a novel notion that we introduce in this work. Our RASMs provide reachability and avoidance guarantees by imposing constraints on what can be viewed as a stochastic extension of level sets of Lyapunov functions for deterministic systems. Our approach solves several important problems -- it can be used to learn a control policy from scratch, to verify a reach-avoid specification for a fixed control policy, or to fine-tune a pre-trained policy if it does not satisfy the reach-avoid specification. We validate our approach on $3$ stochastic non-linear reinforcement learning tasks."}],"ec_funded":1,"date_created":"2023-11-24T13:10:09Z","related_material":{"record":[{"relation":"dissertation_contains","id":"14539","status":"public"},{"relation":"later_version","status":"public","id":"14830"}]},"doi":"10.48550/ARXIV.2210.05308","date_published":"2022-11-29T00:00:00Z","language":[{"iso":"eng"}],"publication":"arXiv","day":"29","publication_status":"submitted","year":"2022","status":"public","project":[{"_id":"0599E47C-7A3F-11EA-A408-12923DDC885E","call_identifier":"H2020","grant_number":"863818","name":"Formal Methods for Stochastic Models: Algorithms and Applications"},{"call_identifier":"H2020","_id":"62781420-2b32-11ec-9570-8d9b63373d4d","name":"Vigilant Algorithmic Monitoring of Software","grant_number":"101020093"},{"call_identifier":"H2020","_id":"2564DBCA-B435-11E9-9278-68D0E5697425","name":"International IST Doctoral Program","grant_number":"665385"}],"tmp":{"short":"CC BY-SA (4.0)","image":"/images/cc_by_sa.png","legal_code_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","name":"Creative Commons Attribution-ShareAlike 4.0 International Public License (CC BY-SA 4.0)"},"type":"preprint","_id":"14600","title":"Learning control policies for stochastic systems with reach-avoid guarantees","department":[{"_id":"KrCh"},{"_id":"ToHe"}],"article_processing_charge":"No","external_id":{"arxiv":["2210.05308"]},"author":[{"full_name":"Zikelic, Dorde","orcid":"0000-0002-4681-1699","last_name":"Zikelic","id":"294AA7A6-F248-11E8-B48F-1D18A9856A87","first_name":"Dorde"},{"id":"3DC22916-F248-11E8-B48F-1D18A9856A87","first_name":"Mathias","full_name":"Lechner, Mathias","last_name":"Lechner"},{"full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724","last_name":"Henzinger","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","first_name":"Thomas A"},{"id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87","first_name":"Krishnendu","last_name":"Chatterjee","full_name":"Chatterjee, Krishnendu","orcid":"0000-0002-4561-241X"}],"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","date_updated":"2024-01-22T14:08:29Z","citation":{"mla":"Zikelic, Dorde, et al. “Learning Control Policies for Stochastic Systems with Reach-Avoid Guarantees.” ArXiv, doi:10.48550/ARXIV.2210.05308.","short":"D. Zikelic, M. Lechner, T.A. Henzinger, K. Chatterjee, ArXiv (n.d.).","ieee":"D. Zikelic, M. Lechner, T. A. Henzinger, and K. Chatterjee, “Learning control policies for stochastic systems with reach-avoid guarantees,” arXiv. .","ama":"Zikelic D, Lechner M, Henzinger TA, Chatterjee K. Learning control policies for stochastic systems with reach-avoid guarantees. arXiv. doi:10.48550/ARXIV.2210.05308","apa":"Zikelic, D., Lechner, M., Henzinger, T. A., & Chatterjee, K. (n.d.). Learning control policies for stochastic systems with reach-avoid guarantees. arXiv. https://doi.org/10.48550/ARXIV.2210.05308","chicago":"Zikelic, Dorde, Mathias Lechner, Thomas A Henzinger, and Krishnendu Chatterjee. “Learning Control Policies for Stochastic Systems with Reach-Avoid Guarantees.” ArXiv, n.d. https://doi.org/10.48550/ARXIV.2210.05308.","ista":"Zikelic D, Lechner M, Henzinger TA, Chatterjee K. Learning control policies for stochastic systems with reach-avoid guarantees. arXiv, 10.48550/ARXIV.2210.05308."}},{"file_date_updated":"2021-10-19T12:52:23Z","department":[{"_id":"ToHe"}],"date_updated":"2021-11-12T11:30:07Z","ddc":["005"],"type":"journal_article","article_type":"original","tmp":{"legal_code_url":"https://creativecommons.org/licenses/by-nd/4.0/legalcode","image":"/image/cc_by_nd.png","name":"Creative Commons Attribution-NoDerivatives 4.0 International (CC BY-ND 4.0)","short":"CC BY-ND (4.0)"},"conference":{"location":"Chicago, IL, United States","end_date":"2021-10-23","start_date":"2021-10-17","name":"OOPSLA: Object-Oriented Programming, Systems, Languages, and Applications"},"status":"public","keyword":["gradual typing","gradual guarantee","nominal","structural","call tags"],"_id":"10153","volume":5,"publication_identifier":{"eissn":["2475-1421"]},"publication_status":"published","file":[{"creator":"fmuehlbo","date_updated":"2021-10-19T12:52:23Z","file_size":770269,"date_created":"2021-10-19T12:52:23Z","file_name":"monnom-oopsla21.pdf","access_level":"open_access","relation":"main_file","content_type":"application/pdf","checksum":"71011efd2da771cafdec7f0d9693f8c1","file_id":"10154","success":1}],"language":[{"iso":"eng"}],"month":"10","intvolume":" 5","abstract":[{"lang":"eng","text":"Gradual typing is a principled means for mixing typed and untyped code. But typed and untyped code often exhibit different programming patterns. There is already substantial research investigating gradually giving types to code exhibiting typical untyped patterns, and some research investigating gradually removing types from code exhibiting typical typed patterns. This paper investigates how to extend these established gradual-typing concepts to give formal guarantees not only about how to change types as code evolves but also about how to change such programming patterns as well.\r\n\r\nIn particular, we explore mixing untyped \"structural\" code with typed \"nominal\" code in an object-oriented language. But whereas previous work only allowed \"nominal\" objects to be treated as \"structural\" objects, we also allow \"structural\" objects to dynamically acquire certain nominal types, namely interfaces. We present a calculus that supports such \"cross-paradigm\" code migration and interoperation in a manner satisfying both the static and dynamic gradual guarantees, and demonstrate that the calculus can be implemented efficiently."}],"oa_version":"Published Version","author":[{"last_name":"Mühlböck","orcid":"0000-0003-1548-0177","full_name":"Mühlböck, Fabian","first_name":"Fabian","id":"6395C5F6-89DF-11E9-9C97-6BDFE5697425"},{"first_name":"Ross","last_name":"Tate","full_name":"Tate, Ross"}],"article_processing_charge":"No","title":"Transitioning from structural to nominal code with efficient gradual typing","citation":{"mla":"Mühlböck, Fabian, and Ross Tate. “Transitioning from Structural to Nominal Code with Efficient Gradual Typing.” Proceedings of the ACM on Programming Languages, vol. 5, 127, Association for Computing Machinery, 2021, doi:10.1145/3485504.","ama":"Mühlböck F, Tate R. Transitioning from structural to nominal code with efficient gradual typing. Proceedings of the ACM on Programming Languages. 2021;5. doi:10.1145/3485504","apa":"Mühlböck, F., & Tate, R. (2021). Transitioning from structural to nominal code with efficient gradual typing. Proceedings of the ACM on Programming Languages. Chicago, IL, United States: Association for Computing Machinery. https://doi.org/10.1145/3485504","short":"F. Mühlböck, R. Tate, Proceedings of the ACM on Programming Languages 5 (2021).","ieee":"F. Mühlböck and R. Tate, “Transitioning from structural to nominal code with efficient gradual typing,” Proceedings of the ACM on Programming Languages, vol. 5. Association for Computing Machinery, 2021.","chicago":"Mühlböck, Fabian, and Ross Tate. “Transitioning from Structural to Nominal Code with Efficient Gradual Typing.” Proceedings of the ACM on Programming Languages. Association for Computing Machinery, 2021. https://doi.org/10.1145/3485504.","ista":"Mühlböck F, Tate R. 2021. Transitioning from structural to nominal code with efficient gradual typing. Proceedings of the ACM on Programming Languages. 5, 127."},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","project":[{"_id":"25F42A32-B435-11E9-9278-68D0E5697425","call_identifier":"FWF","name":"The Wittgenstein Prize","grant_number":"Z211"}],"article_number":"127","date_published":"2021-10-15T00:00:00Z","doi":"10.1145/3485504","date_created":"2021-10-19T12:48:44Z","has_accepted_license":"1","year":"2021","day":"15","publication":"Proceedings of the ACM on Programming Languages","publisher":"Association for Computing Machinery","quality_controlled":"1","oa":1,"acknowledgement":"We thank the reviewers for their valuable suggestions towards improving the paper. We also \r\nthank Mae Milano and Adrian Sampson, as well as the members of the Programming Languages Discussion Group at Cornell University and of the Programming Research Laboratory at Northeastern University, for their helpful feedback on preliminary findings of this work.\r\n\r\nThis material is based upon work supported in part by the National Science Foundation (NSF) through grant CCF-1350182 and the Austrian Science Fund (FWF) through grant Z211-N23 (Wittgenstein~Award).\r\nAny opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF or the FWF."},{"oa":1,"publisher":"AAAI Press","quality_controlled":"1","acknowledgement":"The authors would like to thank the reviewers for their insightful comments. RH and RG were partially supported by\r\nHorizon-2020 ECSEL Project grant No. 783163 (iDev40). RH was partially supported by Boeing. ML was supported\r\nin part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award). SG was funded by FWF\r\nproject W1255-N23. JC was partially supported by NAWA Polish Returns grant PPN/PPO/2018/1/00029. SS was supported by NSF awards DCL-2040599, CCF-1918225, and CPS-1446832.\r\n","date_created":"2022-01-25T15:47:20Z","date_published":"2021-05-28T00:00:00Z","page":"11525-11535","publication":"Proceedings of the AAAI Conference on Artificial Intelligence","day":"28","year":"2021","has_accepted_license":"1","project":[{"name":"The Wittgenstein Prize","grant_number":"Z211","call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425"}],"title":"On the verification of neural ODEs with stochastic guarantees","external_id":{"arxiv":["2012.08863"]},"article_processing_charge":"No","author":[{"first_name":"Sophie","last_name":"Grunbacher","full_name":"Grunbacher, Sophie"},{"first_name":"Ramin","full_name":"Hasani, Ramin","last_name":"Hasani"},{"full_name":"Lechner, Mathias","last_name":"Lechner","first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Cyranka, Jacek","last_name":"Cyranka","first_name":"Jacek"},{"first_name":"Scott A","full_name":"Smolka, Scott A","last_name":"Smolka"},{"first_name":"Radu","last_name":"Grosu","full_name":"Grosu, Radu"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","citation":{"ista":"Grunbacher S, Hasani R, Lechner M, Cyranka J, Smolka SA, Grosu R. 2021. On the verification of neural ODEs with stochastic guarantees. Proceedings of the AAAI Conference on Artificial Intelligence. AAAI: Association for the Advancement of Artificial Intelligence, Technical Tracks, vol. 35, 11525–11535.","chicago":"Grunbacher, Sophie, Ramin Hasani, Mathias Lechner, Jacek Cyranka, Scott A Smolka, and Radu Grosu. “On the Verification of Neural ODEs with Stochastic Guarantees.” In Proceedings of the AAAI Conference on Artificial Intelligence, 35:11525–35. AAAI Press, 2021.","apa":"Grunbacher, S., Hasani, R., Lechner, M., Cyranka, J., Smolka, S. A., & Grosu, R. (2021). On the verification of neural ODEs with stochastic guarantees. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 35, pp. 11525–11535). Virtual: AAAI Press.","ama":"Grunbacher S, Hasani R, Lechner M, Cyranka J, Smolka SA, Grosu R. On the verification of neural ODEs with stochastic guarantees. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol 35. AAAI Press; 2021:11525-11535.","short":"S. Grunbacher, R. Hasani, M. Lechner, J. Cyranka, S.A. Smolka, R. Grosu, in:, Proceedings of the AAAI Conference on Artificial Intelligence, AAAI Press, 2021, pp. 11525–11535.","ieee":"S. Grunbacher, R. Hasani, M. Lechner, J. Cyranka, S. A. Smolka, and R. Grosu, “On the verification of neural ODEs with stochastic guarantees,” in Proceedings of the AAAI Conference on Artificial Intelligence, Virtual, 2021, vol. 35, no. 13, pp. 11525–11535.","mla":"Grunbacher, Sophie, et al. “On the Verification of Neural ODEs with Stochastic Guarantees.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 13, AAAI Press, 2021, pp. 11525–35."},"intvolume":" 35","month":"05","main_file_link":[{"open_access":"1","url":"https://ojs.aaai.org/index.php/AAAI/article/view/17372"}],"alternative_title":["Technical Tracks"],"oa_version":"Published Version","abstract":[{"text":"We show that Neural ODEs, an emerging class of timecontinuous neural networks, can be verified by solving a set of global-optimization problems. For this purpose, we introduce Stochastic Lagrangian Reachability (SLR), an\r\nabstraction-based technique for constructing a tight Reachtube (an over-approximation of the set of reachable states\r\nover a given time-horizon), and provide stochastic guarantees in the form of confidence intervals for the Reachtube bounds. SLR inherently avoids the infamous wrapping effect (accumulation of over-approximation errors) by performing local optimization steps to expand safe regions instead of repeatedly forward-propagating them as is done by deterministic reachability methods. To enable fast local optimizations, we introduce a novel forward-mode adjoint sensitivity method to compute gradients without the need for backpropagation. Finally, we establish asymptotic and non-asymptotic convergence rates for SLR.","lang":"eng"}],"volume":35,"issue":"13","language":[{"iso":"eng"}],"file":[{"date_created":"2022-01-26T07:38:08Z","file_name":"17372-Article Text-20866-1-2-20210518.pdf","creator":"mlechner","date_updated":"2022-01-26T07:38:08Z","file_size":286906,"file_id":"10680","checksum":"468d07041e282a1d46ffdae92f709630","success":1,"access_level":"open_access","relation":"main_file","content_type":"application/pdf"}],"publication_status":"published","publication_identifier":{"issn":["2159-5399"],"isbn":["978-1-57735-866-4"],"eissn":["2374-3468"]},"status":"public","conference":{"name":"AAAI: Association for the Advancement of Artificial Intelligence","location":"Virtual","end_date":"2021-02-09","start_date":"2021-02-02"},"type":"conference","_id":"10669","department":[{"_id":"GradSch"},{"_id":"ToHe"}],"file_date_updated":"2022-01-26T07:38:08Z","ddc":["000"],"date_updated":"2022-05-24T06:33:14Z"},{"publication_identifier":{"issn":["2159-5399"],"eissn":["2374-3468"],"isbn":["978-1-57735-866-4"]},"publication_status":"published","file":[{"content_type":"application/pdf","relation":"main_file","access_level":"open_access","success":1,"file_id":"10678","checksum":"0f06995fba06dbcfa7ed965fc66027ff","file_size":4302669,"date_updated":"2022-01-26T07:36:03Z","creator":"mlechner","file_name":"16936-Article Text-20430-1-2-20210518 (1).pdf","date_created":"2022-01-26T07:36:03Z"}],"language":[{"iso":"eng"}],"issue":"9","volume":35,"abstract":[{"text":"We introduce a new class of time-continuous recurrent neural network models. Instead of declaring a learning system’s dynamics by implicit nonlinearities, we construct networks of linear first-order dynamical systems modulated via nonlinear interlinked gates. The resulting models represent dynamical systems with varying (i.e., liquid) time-constants coupled to their hidden state, with outputs being computed by numerical differential equation solvers. These neural networks exhibit stable and bounded behavior, yield superior expressivity within the family of neural ordinary differential equations, and give rise to improved performance on time-series prediction tasks. To demonstrate these properties, we first take a theoretical approach to find bounds over their dynamics, and compute their expressive power by the trajectory length measure in a latent trajectory space. We then conduct a series of time-series prediction experiments to manifest the approximation capability of Liquid Time-Constant Networks (LTCs) compared to classical and modern RNNs.","lang":"eng"}],"oa_version":"Published Version","alternative_title":["Technical Tracks"],"main_file_link":[{"url":"https://ojs.aaai.org/index.php/AAAI/article/view/16936","open_access":"1"}],"month":"05","intvolume":" 35","date_updated":"2022-05-24T06:36:54Z","ddc":["000"],"file_date_updated":"2022-01-26T07:36:03Z","department":[{"_id":"GradSch"},{"_id":"ToHe"}],"_id":"10671","type":"conference","conference":{"end_date":"2021-02-09","location":"Virtual","start_date":"2021-02-02","name":"AAAI: Association for the Advancement of Artificial Intelligence"},"status":"public","has_accepted_license":"1","year":"2021","day":"28","publication":"Proceedings of the AAAI Conference on Artificial Intelligence","page":"7657-7666","date_published":"2021-05-28T00:00:00Z","date_created":"2022-01-25T15:48:36Z","acknowledgement":"R.H. and D.R. are partially supported by Boeing. R.H. and R.G. were partially supported by the Horizon-2020 ECSEL\r\nProject grant No. 783163 (iDev40). M.L. was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award). A.A. is supported by the National Science Foundation (NSF) Graduate Research Fellowship Program. This research work is partially drawn from the PhD dissertation of R.H.","quality_controlled":"1","publisher":"AAAI Press","oa":1,"citation":{"chicago":"Hasani, Ramin, Mathias Lechner, Alexander Amini, Daniela Rus, and Radu Grosu. “Liquid Time-Constant Networks.” In Proceedings of the AAAI Conference on Artificial Intelligence, 35:7657–66. AAAI Press, 2021.","ista":"Hasani R, Lechner M, Amini A, Rus D, Grosu R. 2021. Liquid time-constant networks. Proceedings of the AAAI Conference on Artificial Intelligence. AAAI: Association for the Advancement of Artificial Intelligence, Technical Tracks, vol. 35, 7657–7666.","mla":"Hasani, Ramin, et al. “Liquid Time-Constant Networks.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 9, AAAI Press, 2021, pp. 7657–66.","ieee":"R. Hasani, M. Lechner, A. Amini, D. Rus, and R. Grosu, “Liquid time-constant networks,” in Proceedings of the AAAI Conference on Artificial Intelligence, Virtual, 2021, vol. 35, no. 9, pp. 7657–7666.","short":"R. Hasani, M. Lechner, A. Amini, D. Rus, R. Grosu, in:, Proceedings of the AAAI Conference on Artificial Intelligence, AAAI Press, 2021, pp. 7657–7666.","apa":"Hasani, R., Lechner, M., Amini, A., Rus, D., & Grosu, R. (2021). Liquid time-constant networks. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 35, pp. 7657–7666). Virtual: AAAI Press.","ama":"Hasani R, Lechner M, Amini A, Rus D, Grosu R. Liquid time-constant networks. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol 35. AAAI Press; 2021:7657-7666."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","author":[{"first_name":"Ramin","full_name":"Hasani, Ramin","last_name":"Hasani"},{"id":"3DC22916-F248-11E8-B48F-1D18A9856A87","first_name":"Mathias","full_name":"Lechner, Mathias","last_name":"Lechner"},{"full_name":"Amini, Alexander","last_name":"Amini","first_name":"Alexander"},{"full_name":"Rus, Daniela","last_name":"Rus","first_name":"Daniela"},{"first_name":"Radu","last_name":"Grosu","full_name":"Grosu, Radu"}],"article_processing_charge":"No","external_id":{"arxiv":["2006.04439"]},"title":"Liquid time-constant networks","project":[{"_id":"25F42A32-B435-11E9-9278-68D0E5697425","call_identifier":"FWF","grant_number":"Z211","name":"The Wittgenstein Prize"}]},{"acknowledgement":"Z.B. is supported by the Doctoral College Resilient Embedded Systems, which is run jointly by the TU Wien’s Faculty of Informatics and the UAS Technikum Wien. R.G. is partially supported by the Horizon 2020 Era-Permed project Persorad, and ECSEL Project grant no. 783163 (iDev40). R.H and D.R were partially supported by Boeing and MIT. M.L. is supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award).","quality_controlled":"1","publisher":"ML Research Press","oa":1,"day":"01","publication":"Proceedings of the 38th International Conference on Machine Learning","has_accepted_license":"1","year":"2021","date_published":"2021-07-01T00:00:00Z","date_created":"2022-01-25T15:46:33Z","page":"478-489","project":[{"grant_number":"Z211","name":"The Wittgenstein Prize","call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425"}],"user_id":"2EBD1598-F248-11E8-B48F-1D18A9856A87","citation":{"mla":"Babaiee, Zahra, et al. “On-off Center-Surround Receptive Fields for Accurate and Robust Image Classification.” Proceedings of the 38th International Conference on Machine Learning, vol. 139, ML Research Press, 2021, pp. 478–89.","short":"Z. Babaiee, R. Hasani, M. Lechner, D. Rus, R. Grosu, in:, Proceedings of the 38th International Conference on Machine Learning, ML Research Press, 2021, pp. 478–489.","ieee":"Z. Babaiee, R. Hasani, M. Lechner, D. Rus, and R. Grosu, “On-off center-surround receptive fields for accurate and robust image classification,” in Proceedings of the 38th International Conference on Machine Learning, Virtual, 2021, vol. 139, pp. 478–489.","apa":"Babaiee, Z., Hasani, R., Lechner, M., Rus, D., & Grosu, R. (2021). On-off center-surround receptive fields for accurate and robust image classification. In Proceedings of the 38th International Conference on Machine Learning (Vol. 139, pp. 478–489). Virtual: ML Research Press.","ama":"Babaiee Z, Hasani R, Lechner M, Rus D, Grosu R. On-off center-surround receptive fields for accurate and robust image classification. In: Proceedings of the 38th International Conference on Machine Learning. Vol 139. ML Research Press; 2021:478-489.","chicago":"Babaiee, Zahra, Ramin Hasani, Mathias Lechner, Daniela Rus, and Radu Grosu. “On-off Center-Surround Receptive Fields for Accurate and Robust Image Classification.” In Proceedings of the 38th International Conference on Machine Learning, 139:478–89. ML Research Press, 2021.","ista":"Babaiee Z, Hasani R, Lechner M, Rus D, Grosu R. 2021. On-off center-surround receptive fields for accurate and robust image classification. Proceedings of the 38th International Conference on Machine Learning. ML: Machine Learning, PMLR, vol. 139, 478–489."},"title":"On-off center-surround receptive fields for accurate and robust image classification","author":[{"first_name":"Zahra","last_name":"Babaiee","full_name":"Babaiee, Zahra"},{"first_name":"Ramin","full_name":"Hasani, Ramin","last_name":"Hasani"},{"first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87","full_name":"Lechner, Mathias","last_name":"Lechner"},{"first_name":"Daniela","last_name":"Rus","full_name":"Rus, Daniela"},{"first_name":"Radu","full_name":"Grosu, Radu","last_name":"Grosu"}],"article_processing_charge":"No","oa_version":"Published Version","abstract":[{"lang":"eng","text":"Robustness to variations in lighting conditions is a key objective for any deep vision system. To this end, our paper extends the receptive field of convolutional neural networks with two residual components, ubiquitous in the visual processing system of vertebrates: On-center and off-center pathways, with an excitatory center and inhibitory surround; OOCS for short. The On-center pathway is excited by the presence of a light stimulus in its center, but not in its surround, whereas the Off-center pathway is excited by the absence of a light stimulus in its center, but not in its surround. We design OOCS pathways via a difference of Gaussians, with their variance computed analytically from the size of the receptive fields. OOCS pathways complement each other in their response to light stimuli, ensuring this way a strong edge-detection capability, and as a result an accurate and robust inference under challenging lighting conditions. We provide extensive empirical evidence showing that networks supplied with OOCS pathways gain accuracy and illumination-robustness from the novel edge representation, compared to other baselines."}],"month":"07","intvolume":" 139","alternative_title":["PMLR"],"main_file_link":[{"open_access":"1","url":"https://proceedings.mlr.press/v139/babaiee21a"}],"file":[{"file_size":4246561,"date_updated":"2022-01-26T07:38:32Z","creator":"mlechner","file_name":"babaiee21a.pdf","date_created":"2022-01-26T07:38:32Z","content_type":"application/pdf","relation":"main_file","access_level":"open_access","success":1,"checksum":"d30eae62561bb517d9f978437d7677db","file_id":"10681"}],"language":[{"iso":"eng"}],"publication_identifier":{"issn":["2640-3498"]},"publication_status":"published","volume":139,"license":"https://creativecommons.org/licenses/by-nc-nd/3.0/","_id":"10668","status":"public","type":"conference","conference":{"end_date":"2021-07-24","location":"Virtual","start_date":"2021-07-18","name":"ML: Machine Learning"},"tmp":{"name":"Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0)","image":"/images/cc_by_nc_nd.png","legal_code_url":"https://creativecommons.org/licenses/by-nc-nd/3.0/legalcode","short":"CC BY-NC-ND (3.0)"},"ddc":["000"],"date_updated":"2022-05-04T15:02:27Z","department":[{"_id":"GradSch"},{"_id":"ToHe"}],"file_date_updated":"2022-01-26T07:38:32Z"},{"type":"conference","tmp":{"name":"Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0)","image":"/images/cc_by_nc_nd.png","legal_code_url":"https://creativecommons.org/licenses/by-nc-nd/3.0/legalcode","short":"CC BY-NC-ND (3.0)"},"conference":{"start_date":"2021-12-06","location":"Virtual","end_date":"2021-12-10","name":"NeurIPS: Neural Information Processing Systems"},"status":"public","_id":"10670","department":[{"_id":"GradSch"},{"_id":"ToHe"}],"file_date_updated":"2022-01-26T07:37:24Z","date_updated":"2022-01-26T14:33:31Z","ddc":["000"],"alternative_title":[" Advances in Neural Information Processing Systems"],"main_file_link":[{"url":"https://proceedings.neurips.cc/paper/2021/hash/67ba02d73c54f0b83c05507b7fb7267f-Abstract.html","open_access":"1"}],"month":"12","abstract":[{"text":"Imitation learning enables high-fidelity, vision-based learning of policies within rich, photorealistic environments. However, such techniques often rely on traditional discrete-time neural models and face difficulties in generalizing to domain shifts by failing to account for the causal relationships between the agent and the environment. In this paper, we propose a theoretical and experimental framework for learning causal representations using continuous-time neural networks, specifically over their discrete-time counterparts. We evaluate our method in the context of visual-control learning of drones over a series of complex tasks, ranging from short- and long-term navigation, to chasing static and dynamic objects through photorealistic environments. Our results demonstrate that causal continuous-time\r\ndeep models can perform robust navigation tasks, where advanced recurrent models fail. These models learn complex causal control representations directly from raw visual inputs and scale to solve a variety of tasks using imitation learning.","lang":"eng"}],"oa_version":"Published Version","publication_status":"published","file":[{"file_id":"10679","checksum":"be81f0ade174a8c9b2d4fe09590b2021","success":1,"access_level":"open_access","relation":"main_file","content_type":"application/pdf","date_created":"2022-01-26T07:37:24Z","file_name":"NeurIPS-2021-causal-navigation-by-continuous-time-neural-networks-Paper.pdf","creator":"mlechner","date_updated":"2022-01-26T07:37:24Z","file_size":6841228}],"language":[{"iso":"eng"}],"project":[{"grant_number":"Z211","name":"The Wittgenstein Prize","call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425"}],"author":[{"last_name":"Vorbach","full_name":"Vorbach, Charles J","first_name":"Charles J"},{"last_name":"Hasani","full_name":"Hasani, Ramin","first_name":"Ramin"},{"last_name":"Amini","full_name":"Amini, Alexander","first_name":"Alexander"},{"first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87","last_name":"Lechner","full_name":"Lechner, Mathias"},{"full_name":"Rus, Daniela","last_name":"Rus","first_name":"Daniela"}],"external_id":{"arxiv":["2106.08314"]},"article_processing_charge":"No","title":"Causal navigation by continuous-time neural networks","citation":{"mla":"Vorbach, Charles J., et al. “Causal Navigation by Continuous-Time Neural Networks.” 35th Conference on Neural Information Processing Systems, 2021.","ama":"Vorbach CJ, Hasani R, Amini A, Lechner M, Rus D. Causal navigation by continuous-time neural networks. In: 35th Conference on Neural Information Processing Systems. ; 2021.","apa":"Vorbach, C. J., Hasani, R., Amini, A., Lechner, M., & Rus, D. (2021). Causal navigation by continuous-time neural networks. In 35th Conference on Neural Information Processing Systems. Virtual.","ieee":"C. J. Vorbach, R. Hasani, A. Amini, M. Lechner, and D. Rus, “Causal navigation by continuous-time neural networks,” in 35th Conference on Neural Information Processing Systems, Virtual, 2021.","short":"C.J. Vorbach, R. Hasani, A. Amini, M. Lechner, D. Rus, in:, 35th Conference on Neural Information Processing Systems, 2021.","chicago":"Vorbach, Charles J, Ramin Hasani, Alexander Amini, Mathias Lechner, and Daniela Rus. “Causal Navigation by Continuous-Time Neural Networks.” In 35th Conference on Neural Information Processing Systems, 2021.","ista":"Vorbach CJ, Hasani R, Amini A, Lechner M, Rus D. 2021. Causal navigation by continuous-time neural networks. 35th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, ."},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","quality_controlled":"1","oa":1,"acknowledgement":"C.V., R.H. A.A. and D.R. are partially supported by Boeing and MIT. A.A. is supported by the National Science Foundation (NSF) Graduate Research Fellowship Program. M.L. is supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award). Research was sponsored by the United States Air Force Research Laboratory and the United States Air Force Artificial Intelligence Accelerator and was accomplished under Cooperative Agreement Number FA8750-19-2-1000. The views and conclusions contained in this document are those of the authors\r\nand should not be interpreted as representing the official policies, either expressed or implied, of the United States Air Force or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.\r\n","date_published":"2021-12-01T00:00:00Z","date_created":"2022-01-25T15:47:50Z","has_accepted_license":"1","year":"2021","day":"01","publication":"35th Conference on Neural Information Processing Systems"},{"project":[{"call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425","name":"The Wittgenstein Prize","grant_number":"Z211"}],"citation":{"mla":"Kragl, Bernhard, and Shaz Qadeer. “The Civl Verifier.” Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design, edited by Piskac Ruzica and Michael W. Whalen, vol. 2, TU Wien Academic Press, 2021, pp. 143–152, doi:10.34727/2021/isbn.978-3-85448-046-4_23.","apa":"Kragl, B., & Qadeer, S. (2021). The Civl verifier. In P. Ruzica & M. W. Whalen (Eds.), Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design (Vol. 2, pp. 143–152). Virtual: TU Wien Academic Press. https://doi.org/10.34727/2021/isbn.978-3-85448-046-4_23","ama":"Kragl B, Qadeer S. The Civl verifier. In: Ruzica P, Whalen MW, eds. Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design. Vol 2. TU Wien Academic Press; 2021:143–152. doi:10.34727/2021/isbn.978-3-85448-046-4_23","short":"B. Kragl, S. Qadeer, in:, P. Ruzica, M.W. Whalen (Eds.), Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design, TU Wien Academic Press, 2021, pp. 143–152.","ieee":"B. Kragl and S. Qadeer, “The Civl verifier,” in Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design, Virtual, 2021, vol. 2, pp. 143–152.","chicago":"Kragl, Bernhard, and Shaz Qadeer. “The Civl Verifier.” In Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design, edited by Piskac Ruzica and Michael W. Whalen, 2:143–152. TU Wien Academic Press, 2021. https://doi.org/10.34727/2021/isbn.978-3-85448-046-4_23.","ista":"Kragl B, Qadeer S. 2021. The Civl verifier. Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design. FMCAD: Formal Methods in Computer-Aided Design, Conference Series, vol. 2, 143–152."},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","author":[{"last_name":"Kragl","full_name":"Kragl, Bernhard","orcid":"0000-0001-7745-9117","first_name":"Bernhard","id":"320FC952-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Shaz","last_name":"Qadeer","full_name":"Qadeer, Shaz"}],"article_processing_charge":"No","title":"The Civl verifier","editor":[{"first_name":"Piskac","last_name":"Ruzica","full_name":"Ruzica, Piskac"},{"full_name":"Whalen, Michael W.","last_name":"Whalen","first_name":"Michael W."}],"acknowledgement":"This research was performed while Bernhard Kragl was at IST Austria, supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award).","quality_controlled":"1","publisher":"TU Wien Academic Press","oa":1,"has_accepted_license":"1","year":"2021","day":"01","publication":"Proceedings of the 21st Conference on Formal Methods in Computer-Aided Design","page":"143–152","date_published":"2021-10-01T00:00:00Z","doi":"10.34727/2021/isbn.978-3-85448-046-4_23","date_created":"2022-01-26T08:01:30Z","_id":"10688","type":"conference","conference":{"start_date":"2021-10-20","location":"Virtual","end_date":"2021-10-22","name":"FMCAD: Formal Methods in Computer-Aided Design"},"tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)"},"status":"public","date_updated":"2022-01-26T08:20:41Z","ddc":["000"],"department":[{"_id":"ToHe"}],"file_date_updated":"2022-01-26T08:04:29Z","abstract":[{"text":"Civl is a static verifier for concurrent programs designed around the conceptual framework of layered refinement,\r\nwhich views the task of verifying a program as a sequence of program simplification steps each justified by its own invariant. Civl verifies a layered concurrent program that compactly expresses all the programs in this sequence and the supporting invariants. This paper presents the design and implementation of the Civl verifier.","lang":"eng"}],"oa_version":"Published Version","alternative_title":["Conference Series"],"month":"10","intvolume":" 2","publication_identifier":{"isbn":["978-3-85448-046-4"]},"publication_status":"published","file":[{"creator":"cchlebak","file_size":390555,"date_updated":"2022-01-26T08:04:29Z","file_name":"2021_FCAD2021_Kragl.pdf","date_created":"2022-01-26T08:04:29Z","relation":"main_file","access_level":"open_access","content_type":"application/pdf","success":1,"file_id":"10689","checksum":"35438ac9f9750340b7f8ae4ae3220d9f"}],"language":[{"iso":"eng"}],"volume":2},{"_id":"9281","article_number":"2103.11389","type":"preprint","project":[{"_id":"260C2330-B435-11E9-9278-68D0E5697425","call_identifier":"H2020","name":"ISTplus - Postdoctoral Fellowships","grant_number":"754411"}],"status":"public","date_updated":"2023-05-03T10:26:45Z","citation":{"chicago":"Dubach, Guillaume, and Fabian Mühlböck. “Formal Verification of Zagier’s One-Sentence Proof.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2103.11389.","ista":"Dubach G, Mühlböck F. Formal verification of Zagier’s one-sentence proof. arXiv, 2103.11389.","mla":"Dubach, Guillaume, and Fabian Mühlböck. “Formal Verification of Zagier’s One-Sentence Proof.” ArXiv, 2103.11389, doi:10.48550/arXiv.2103.11389.","ama":"Dubach G, Mühlböck F. Formal verification of Zagier’s one-sentence proof. arXiv. doi:10.48550/arXiv.2103.11389","apa":"Dubach, G., & Mühlböck, F. (n.d.). Formal verification of Zagier’s one-sentence proof. arXiv. https://doi.org/10.48550/arXiv.2103.11389","short":"G. Dubach, F. Mühlböck, ArXiv (n.d.).","ieee":"G. Dubach and F. Mühlböck, “Formal verification of Zagier’s one-sentence proof,” arXiv. ."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","author":[{"id":"D5C6A458-10C4-11EA-ABF4-A4B43DDC885E","first_name":"Guillaume","last_name":"Dubach","full_name":"Dubach, Guillaume","orcid":"0000-0001-6892-8137"},{"last_name":"Mühlböck","full_name":"Mühlböck, Fabian","orcid":"0000-0003-1548-0177","first_name":"Fabian","id":"6395C5F6-89DF-11E9-9C97-6BDFE5697425"}],"article_processing_charge":"No","external_id":{"arxiv":["2103.11389"]},"title":"Formal verification of Zagier's one-sentence proof","department":[{"_id":"LaEr"},{"_id":"ToHe"}],"abstract":[{"text":"We comment on two formal proofs of Fermat's sum of two squares theorem, written using the Mathematical Components libraries of the Coq proof assistant. The first one follows Zagier's celebrated one-sentence proof; the second follows David Christopher's recent new proof relying on partition-theoretic arguments. Both formal proofs rely on a general property of involutions of finite sets, of independent interest. The proof technique consists for the most part of automating recurrent tasks (such as case distinctions and computations on natural numbers) via ad hoc tactics.","lang":"eng"}],"oa_version":"Preprint","main_file_link":[{"url":"https://arxiv.org/abs/2103.11389","open_access":"1"}],"oa":1,"month":"03","year":"2021","publication_status":"submitted","day":"21","language":[{"iso":"eng"}],"publication":"arXiv","related_material":{"record":[{"relation":"other","id":"9946","status":"public"}]},"doi":"10.48550/arXiv.2103.11389","date_published":"2021-03-21T00:00:00Z","date_created":"2021-03-23T05:38:48Z","ec_funded":1},{"publication_identifier":{"isbn":["978-1-57735-866-4"],"eissn":["2374-3468"],"issn":["2159-5399"]},"publication_status":"published","file":[{"content_type":"application/pdf","access_level":"open_access","relation":"main_file","checksum":"2bc8155b2526a70fba5b7301bc89dbd1","file_id":"10684","success":1,"date_updated":"2022-01-26T07:41:16Z","file_size":137235,"creator":"mlechner","date_created":"2022-01-26T07:41:16Z","file_name":"16496-Article Text-19990-1-2-20210518 (1).pdf"}],"language":[{"iso":"eng"}],"volume":35,"related_material":{"record":[{"relation":"dissertation_contains","status":"public","id":"11362"}]},"issue":"5A","ec_funded":1,"abstract":[{"text":"Formal verification of neural networks is an active topic of research, and recent advances have significantly increased the size of the networks that verification tools can handle. However, most methods are designed for verification of an idealized model of the actual network which works over real arithmetic and ignores rounding imprecisions. This idealization is in stark contrast to network quantization, which is a technique that trades numerical precision for computational efficiency and is, therefore, often applied in practice. Neglecting rounding errors of such low-bit quantized neural networks has been shown to lead to wrong conclusions about the network’s correctness. Thus, the desired approach for verifying quantized neural networks would be one that takes these rounding errors\r\ninto account. In this paper, we show that verifying the bitexact implementation of quantized neural networks with bitvector specifications is PSPACE-hard, even though verifying idealized real-valued networks and satisfiability of bit-vector specifications alone are each in NP. Furthermore, we explore several practical heuristics toward closing the complexity gap between idealized and bit-exact verification. In particular, we propose three techniques for making SMT-based verification of quantized neural networks more scalable. Our experiments demonstrate that our proposed methods allow a speedup of up to three orders of magnitude over existing approaches.","lang":"eng"}],"oa_version":"Published Version","alternative_title":["Technical Tracks"],"scopus_import":"1","main_file_link":[{"url":"https://ojs.aaai.org/index.php/AAAI/article/view/16496","open_access":"1"}],"month":"05","intvolume":" 35","date_updated":"2023-06-23T07:01:11Z","ddc":["000"],"department":[{"_id":"GradSch"},{"_id":"ToHe"}],"file_date_updated":"2022-01-26T07:41:16Z","_id":"10665","type":"conference","conference":{"name":"AAAI: Association for the Advancement of Artificial Intelligence","location":"Virtual","end_date":"2021-02-09","start_date":"2021-02-02"},"status":"public","has_accepted_license":"1","year":"2021","day":"28","publication":"Proceedings of the AAAI Conference on Artificial Intelligence","page":"3787-3795","date_published":"2021-05-28T00:00:00Z","date_created":"2022-01-25T15:15:02Z","acknowledgement":"This research was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein\r\nAward), ERC CoG 863818 (FoRM-SMArt), and the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie Grant Agreement No. 665385.\r\n","quality_controlled":"1","publisher":"AAAI Press","oa":1,"citation":{"ista":"Henzinger TA, Lechner M, Zikelic D. 2021. Scalable verification of quantized neural networks. Proceedings of the AAAI Conference on Artificial Intelligence. AAAI: Association for the Advancement of Artificial Intelligence, Technical Tracks, vol. 35, 3787–3795.","chicago":"Henzinger, Thomas A, Mathias Lechner, and Dorde Zikelic. “Scalable Verification of Quantized Neural Networks.” In Proceedings of the AAAI Conference on Artificial Intelligence, 35:3787–95. AAAI Press, 2021.","ieee":"T. A. Henzinger, M. Lechner, and D. Zikelic, “Scalable verification of quantized neural networks,” in Proceedings of the AAAI Conference on Artificial Intelligence, Virtual, 2021, vol. 35, no. 5A, pp. 3787–3795.","short":"T.A. Henzinger, M. Lechner, D. Zikelic, in:, Proceedings of the AAAI Conference on Artificial Intelligence, AAAI Press, 2021, pp. 3787–3795.","ama":"Henzinger TA, Lechner M, Zikelic D. Scalable verification of quantized neural networks. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol 35. AAAI Press; 2021:3787-3795.","apa":"Henzinger, T. A., Lechner, M., & Zikelic, D. (2021). Scalable verification of quantized neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 35, pp. 3787–3795). Virtual: AAAI Press.","mla":"Henzinger, Thomas A., et al. “Scalable Verification of Quantized Neural Networks.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 5A, AAAI Press, 2021, pp. 3787–95."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","author":[{"last_name":"Henzinger","orcid":"0000-0002-2985-7724","full_name":"Henzinger, Thomas A","first_name":"Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87","last_name":"Lechner","full_name":"Lechner, Mathias"},{"id":"294AA7A6-F248-11E8-B48F-1D18A9856A87","first_name":"Dorde","last_name":"Zikelic","full_name":"Zikelic, Dorde"}],"external_id":{"arxiv":["2012.08185"]},"article_processing_charge":"No","title":"Scalable verification of quantized neural networks","project":[{"call_identifier":"H2020","_id":"2564DBCA-B435-11E9-9278-68D0E5697425","grant_number":"665385","name":"International IST Doctoral Program"},{"call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425","name":"The Wittgenstein Prize","grant_number":"Z211"},{"name":"Formal Methods for Stochastic Models: Algorithms and Applications","grant_number":"863818","_id":"0599E47C-7A3F-11EA-A408-12923DDC885E","call_identifier":"H2020"}]},{"project":[{"call_identifier":"H2020","_id":"2564DBCA-B435-11E9-9278-68D0E5697425","name":"International IST Doctoral Program","grant_number":"665385"},{"_id":"0599E47C-7A3F-11EA-A408-12923DDC885E","call_identifier":"H2020","grant_number":"863818","name":"Formal Methods for Stochastic Models: Algorithms and Applications"},{"call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425","name":"The Wittgenstein Prize","grant_number":"Z211"}],"citation":{"mla":"Lechner, Mathias, et al. “Infinite Time Horizon Safety of Bayesian Neural Networks.” 35th Conference on Neural Information Processing Systems, 2021, doi:10.48550/arXiv.2111.03165.","ieee":"M. Lechner, Ð. Žikelić, K. Chatterjee, and T. A. Henzinger, “Infinite time horizon safety of Bayesian neural networks,” in 35th Conference on Neural Information Processing Systems, Virtual, 2021.","short":"M. Lechner, Ð. Žikelić, K. Chatterjee, T.A. Henzinger, in:, 35th Conference on Neural Information Processing Systems, 2021.","ama":"Lechner M, Žikelić Ð, Chatterjee K, Henzinger TA. Infinite time horizon safety of Bayesian neural networks. In: 35th Conference on Neural Information Processing Systems. ; 2021. doi:10.48550/arXiv.2111.03165","apa":"Lechner, M., Žikelić, Ð., Chatterjee, K., & Henzinger, T. A. (2021). Infinite time horizon safety of Bayesian neural networks. In 35th Conference on Neural Information Processing Systems. Virtual. https://doi.org/10.48550/arXiv.2111.03165","chicago":"Lechner, Mathias, Ðorđe Žikelić, Krishnendu Chatterjee, and Thomas A Henzinger. “Infinite Time Horizon Safety of Bayesian Neural Networks.” In 35th Conference on Neural Information Processing Systems, 2021. https://doi.org/10.48550/arXiv.2111.03165.","ista":"Lechner M, Žikelić Ð, Chatterjee K, Henzinger TA. 2021. Infinite time horizon safety of Bayesian neural networks. 35th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, ."},"user_id":"2EBD1598-F248-11E8-B48F-1D18A9856A87","article_processing_charge":"No","external_id":{"arxiv":["2111.03165"]},"author":[{"id":"3DC22916-F248-11E8-B48F-1D18A9856A87","first_name":"Mathias","last_name":"Lechner","full_name":"Lechner, Mathias"},{"last_name":"Žikelić","full_name":"Žikelić, Ðorđe","first_name":"Ðorđe"},{"first_name":"Krishnendu","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87","full_name":"Chatterjee, Krishnendu","orcid":"0000-0002-4561-241X","last_name":"Chatterjee"},{"full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724","last_name":"Henzinger","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","first_name":"Thomas A"}],"title":"Infinite time horizon safety of Bayesian neural networks","acknowledgement":"This research was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award), ERC CoG 863818 (FoRM-SMArt), and the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie Grant Agreement No. 665385.","oa":1,"quality_controlled":"1","year":"2021","has_accepted_license":"1","publication":"35th Conference on Neural Information Processing Systems","day":"01","date_created":"2022-01-25T15:45:58Z","date_published":"2021-12-01T00:00:00Z","doi":"10.48550/arXiv.2111.03165","_id":"10667","conference":{"start_date":"2021-12-06","location":"Virtual","end_date":"2021-12-10","name":"NeurIPS: Neural Information Processing Systems"},"tmp":{"name":"Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0)","image":"/images/cc_by_nc_nd.png","legal_code_url":"https://creativecommons.org/licenses/by-nc-nd/3.0/legalcode","short":"CC BY-NC-ND (3.0)"},"type":"conference","status":"public","date_updated":"2023-06-23T07:01:11Z","ddc":["000"],"file_date_updated":"2022-01-26T07:39:59Z","department":[{"_id":"GradSch"},{"_id":"ToHe"},{"_id":"KrCh"}],"abstract":[{"text":"Bayesian neural networks (BNNs) place distributions over the weights of a neural network to model uncertainty in the data and the network's prediction. We consider the problem of verifying safety when running a Bayesian neural network policy in a feedback loop with infinite time horizon systems. Compared to the existing sampling-based approaches, which are inapplicable to the infinite time horizon setting, we train a separate deterministic neural network that serves as an infinite time horizon safety certificate. In particular, we show that the certificate network guarantees the safety of the system over a subset of the BNN weight posterior's support. Our method first computes a safe weight set and then alters the BNN's weight posterior to reject samples outside this set. Moreover, we show how to extend our approach to a safe-exploration reinforcement learning setting, in order to avoid unsafe trajectories during the training of the policy. We evaluate our approach on a series of reinforcement learning benchmarks, including non-Lyapunovian safety specifications.","lang":"eng"}],"oa_version":"Published Version","main_file_link":[{"open_access":"1","url":"https://proceedings.neurips.cc/paper/2021/hash/544defa9fddff50c53b71c43e0da72be-Abstract.html"}],"alternative_title":[" Advances in Neural Information Processing Systems"],"month":"12","publication_status":"published","language":[{"iso":"eng"}],"file":[{"success":1,"checksum":"0fc0f852525c10dda9cc9ffea07fb4e4","file_id":"10682","relation":"main_file","access_level":"open_access","content_type":"application/pdf","file_name":"infinite_time_horizon_safety_o.pdf","date_created":"2022-01-26T07:39:59Z","creator":"mlechner","file_size":452492,"date_updated":"2022-01-26T07:39:59Z"}],"ec_funded":1,"related_material":{"record":[{"status":"public","id":"11362","relation":"dissertation_contains"}]}},{"department":[{"_id":"ToHe"}],"file_date_updated":"2020-12-02T13:33:51Z","date_updated":"2023-08-04T11:19:00Z","ddc":["000"],"type":"journal_article","article_type":"original","status":"public","_id":"8912","volume":167,"issue":"4","publication_status":"published","publication_identifier":{"issn":["09574174"]},"language":[{"iso":"eng"}],"file":[{"file_size":634967,"date_updated":"2020-12-02T13:33:51Z","creator":"esarac","file_name":"synchroPaperRevised.pdf","date_created":"2020-12-02T13:33:51Z","content_type":"application/pdf","relation":"main_file","access_level":"open_access","file_id":"8913","checksum":"600c2f81bc898a725bcfa7cf26ff4fed"}],"scopus_import":"1","intvolume":" 167","month":"04","abstract":[{"text":"For automata, synchronization, the problem of bringing an automaton to a particular state regardless of its initial state, is important. It has several applications in practice and is related to a fifty-year-old conjecture on the length of the shortest synchronizing word. Although using shorter words increases the effectiveness in practice, finding a shortest one (which is not necessarily unique) is NP-hard. For this reason, there exist various heuristics in the literature. However, high-quality heuristics such as SynchroP producing relatively shorter sequences are very expensive and can take hours when the automaton has tens of thousands of states. The SynchroP heuristic has been frequently used as a benchmark to evaluate the performance of the new heuristics. In this work, we first improve the runtime of SynchroP and its variants by using algorithmic techniques. We then focus on adapting SynchroP for many-core architectures,\r\nand overall, we obtain more than 1000× speedup on GPUs compared to naive sequential implementation that has been frequently used as a benchmark to evaluate new heuristics in the literature. We also propose two SynchroP variants and evaluate their performance.","lang":"eng"}],"oa_version":"Submitted Version","article_processing_charge":"No","external_id":{"isi":["000640531100038"]},"author":[{"full_name":"Sarac, Naci E","last_name":"Sarac","id":"8C6B42F8-C8E6-11E9-A03A-F2DCE5697425","first_name":"Naci E"},{"full_name":"Altun, Ömer Faruk","last_name":"Altun","first_name":"Ömer Faruk"},{"first_name":"Kamil Tolga","last_name":"Atam","full_name":"Atam, Kamil Tolga"},{"last_name":"Karahoda","full_name":"Karahoda, Sertac","first_name":"Sertac"},{"last_name":"Kaya","full_name":"Kaya, Kamer","first_name":"Kamer"},{"first_name":"Hüsnü","full_name":"Yenigün, Hüsnü","last_name":"Yenigün"}],"title":"Boosting expensive synchronizing heuristics","citation":{"chicago":"Sarac, Naci E, Ömer Faruk Altun, Kamil Tolga Atam, Sertac Karahoda, Kamer Kaya, and Hüsnü Yenigün. “Boosting Expensive Synchronizing Heuristics.” Expert Systems with Applications. Elsevier, 2021. https://doi.org/10.1016/j.eswa.2020.114203.","ista":"Sarac NE, Altun ÖF, Atam KT, Karahoda S, Kaya K, Yenigün H. 2021. Boosting expensive synchronizing heuristics. Expert Systems with Applications. 167(4), 114203.","mla":"Sarac, Naci E., et al. “Boosting Expensive Synchronizing Heuristics.” Expert Systems with Applications, vol. 167, no. 4, 114203, Elsevier, 2021, doi:10.1016/j.eswa.2020.114203.","apa":"Sarac, N. E., Altun, Ö. F., Atam, K. T., Karahoda, S., Kaya, K., & Yenigün, H. (2021). Boosting expensive synchronizing heuristics. Expert Systems with Applications. Elsevier. https://doi.org/10.1016/j.eswa.2020.114203","ama":"Sarac NE, Altun ÖF, Atam KT, Karahoda S, Kaya K, Yenigün H. Boosting expensive synchronizing heuristics. Expert Systems with Applications. 2021;167(4). doi:10.1016/j.eswa.2020.114203","short":"N.E. Sarac, Ö.F. Altun, K.T. Atam, S. Karahoda, K. Kaya, H. Yenigün, Expert Systems with Applications 167 (2021).","ieee":"N. E. Sarac, Ö. F. Altun, K. T. Atam, S. Karahoda, K. Kaya, and H. Yenigün, “Boosting expensive synchronizing heuristics,” Expert Systems with Applications, vol. 167, no. 4. Elsevier, 2021."},"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","project":[{"name":"The Wittgenstein Prize","grant_number":"Z211","_id":"25F42A32-B435-11E9-9278-68D0E5697425","call_identifier":"FWF"}],"article_number":"114203","date_created":"2020-12-02T13:34:25Z","date_published":"2021-04-01T00:00:00Z","doi":"10.1016/j.eswa.2020.114203","year":"2021","isi":1,"has_accepted_license":"1","publication":"Expert Systems with Applications","day":"01","oa":1,"publisher":"Elsevier","quality_controlled":"1","acknowledgement":"This work was supported by The Scientific and Technological Research Council of Turkey (TUBITAK) [grant number 114E569]. This research was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award). We would like to thank the authors of (Roman & Szykula, 2015) for providing their heuristics implementations, which we used to compare our SynchroP implementation as given in Table 11."},{"scopus_import":"1","month":"05","abstract":[{"text":"Formal design of embedded and cyber-physical systems relies on mathematical modeling. In this paper, we consider the model class of hybrid automata whose dynamics are defined by affine differential equations. Given a set of time-series data, we present an algorithmic approach to synthesize a hybrid automaton exhibiting behavior that is close to the data, up to a specified precision, and changes in synchrony with the data. A fundamental problem in our synthesis algorithm is to check membership of a time series in a hybrid automaton. Our solution integrates reachability and optimization techniques for affine dynamical systems to obtain both a sufficient and a necessary condition for membership, combined in a refinement framework. The algorithm processes one time series at a time and hence can be interrupted, provide an intermediate result, and be resumed. We report experimental results demonstrating the applicability of our synthesis approach.","lang":"eng"}],"oa_version":"Published Version","ec_funded":1,"publication_identifier":{"isbn":["9781450383394"]},"publication_status":"published","file":[{"file_size":1474786,"date_updated":"2021-05-25T13:53:22Z","creator":"kschuh","file_name":"2021_HSCC_Soto.pdf","date_created":"2021-05-25T13:53:22Z","content_type":"application/pdf","relation":"main_file","access_level":"open_access","success":1,"file_id":"9424","checksum":"4c1202c1abf71384c3ee6fea88c2f80e"}],"language":[{"iso":"eng"}],"type":"conference","conference":{"name":"HSCC: International Conference on Hybrid Systems Computation and Control","start_date":"2021-05-19","location":"Nashville, TN, United States","end_date":"2021-05-21"},"tmp":{"legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","image":"/images/cc_by.png","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","short":"CC BY (4.0)"},"status":"public","keyword":["hybrid automaton","membership","system identification"],"_id":"9200","department":[{"_id":"ToHe"}],"file_date_updated":"2021-05-25T13:53:22Z","date_updated":"2023-08-07T13:49:33Z","ddc":["000"],"publisher":"Association for Computing Machinery","quality_controlled":"1","oa":1,"acknowledgement":"This research was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award) and the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 754411.","page":"2102.12734","date_published":"2021-05-01T00:00:00Z","doi":"10.1145/3447928.3456704","date_created":"2021-02-26T16:30:39Z","has_accepted_license":"1","isi":1,"year":"2021","day":"01","publication":"HSCC '21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control","project":[{"name":"The Wittgenstein Prize","grant_number":"Z211","call_identifier":"FWF","_id":"25F42A32-B435-11E9-9278-68D0E5697425"},{"_id":"260C2330-B435-11E9-9278-68D0E5697425","call_identifier":"H2020","name":"ISTplus - Postdoctoral Fellowships","grant_number":"754411"}],"author":[{"id":"4B3207F6-F248-11E8-B48F-1D18A9856A87","first_name":"Miriam","last_name":"Garcia Soto","orcid":"0000-0003-2936-5719","full_name":"Garcia Soto, Miriam"},{"id":"40876CD8-F248-11E8-B48F-1D18A9856A87","first_name":"Thomas A","orcid":"0000-0002-2985-7724","full_name":"Henzinger, Thomas A","last_name":"Henzinger"},{"orcid":"0000-0003-3658-1065","full_name":"Schilling, Christian","last_name":"Schilling","first_name":"Christian","id":"3A2F4DCE-F248-11E8-B48F-1D18A9856A87"}],"external_id":{"isi":["000932821700028"],"arxiv":["2102.12734"]},"article_processing_charge":"No","title":"Synthesis of hybrid automata with affine dynamics from time-series data","citation":{"ieee":"M. Garcia Soto, T. A. Henzinger, and C. Schilling, “Synthesis of hybrid automata with affine dynamics from time-series data,” in HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control, Nashville, TN, United States, 2021, p. 2102.12734.","short":"M. Garcia Soto, T.A. Henzinger, C. Schilling, in:, HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control, Association for Computing Machinery, 2021, p. 2102.12734.","apa":"Garcia Soto, M., Henzinger, T. A., & Schilling, C. (2021). Synthesis of hybrid automata with affine dynamics from time-series data. In HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control (p. 2102.12734). Nashville, TN, United States: Association for Computing Machinery. https://doi.org/10.1145/3447928.3456704","ama":"Garcia Soto M, Henzinger TA, Schilling C. Synthesis of hybrid automata with affine dynamics from time-series data. In: HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control. Association for Computing Machinery; 2021:2102.12734. doi:10.1145/3447928.3456704","mla":"Garcia Soto, Miriam, et al. “Synthesis of Hybrid Automata with Affine Dynamics from Time-Series Data.” HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control, Association for Computing Machinery, 2021, p. 2102.12734, doi:10.1145/3447928.3456704.","ista":"Garcia Soto M, Henzinger TA, Schilling C. 2021. Synthesis of hybrid automata with affine dynamics from time-series data. HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control. HSCC: International Conference on Hybrid Systems Computation and Control, 2102.12734.","chicago":"Garcia Soto, Miriam, Thomas A Henzinger, and Christian Schilling. “Synthesis of Hybrid Automata with Affine Dynamics from Time-Series Data.” In HSCC ’21: Proceedings of the 24th International Conference on Hybrid Systems: Computation and Control, 2102.12734. Association for Computing Machinery, 2021. https://doi.org/10.1145/3447928.3456704."},"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8"},{"title":"Bidding mechanisms in graph games","article_processing_charge":"No","external_id":{"isi":["000634149800009"],"arxiv":["1905.03835"]},"author":[{"first_name":"Guy","id":"463C8BC2-F248-11E8-B48F-1D18A9856A87","full_name":"Avni, Guy","orcid":"0000-0001-5588-8287","last_name":"Avni"},{"last_name":"Henzinger","full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724","first_name":"Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Đorđe","last_name":"Žikelić","full_name":"Žikelić, Đorđe"}],"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","citation":{"chicago":"Avni, Guy, Thomas A Henzinger, and Đorđe Žikelić. “Bidding Mechanisms in Graph Games.” Journal of Computer and System Sciences. Elsevier, 2021. https://doi.org/10.1016/j.jcss.2021.02.008.","ista":"Avni G, Henzinger TA, Žikelić Đ. 2021. Bidding mechanisms in graph games. Journal of Computer and System Sciences. 119(8), 133–144.","mla":"Avni, Guy, et al. “Bidding Mechanisms in Graph Games.” Journal of Computer and System Sciences, vol. 119, no. 8, Elsevier, 2021, pp. 133–44, doi:10.1016/j.jcss.2021.02.008.","apa":"Avni, G., Henzinger, T. A., & Žikelić, Đ. (2021). Bidding mechanisms in graph games. Journal of Computer and System Sciences. Elsevier. https://doi.org/10.1016/j.jcss.2021.02.008","ama":"Avni G, Henzinger TA, Žikelić Đ. Bidding mechanisms in graph games. Journal of Computer and System Sciences. 2021;119(8):133-144. doi:10.1016/j.jcss.2021.02.008","ieee":"G. Avni, T. A. Henzinger, and Đ. Žikelić, “Bidding mechanisms in graph games,” Journal of Computer and System Sciences, vol. 119, no. 8. Elsevier, pp. 133–144, 2021.","short":"G. Avni, T.A. Henzinger, Đ. Žikelić, Journal of Computer and System Sciences 119 (2021) 133–144."},"oa":1,"quality_controlled":"1","publisher":"Elsevier","date_created":"2021-03-14T23:01:32Z","date_published":"2021-03-03T00:00:00Z","doi":"10.1016/j.jcss.2021.02.008","page":"133-144","publication":"Journal of Computer and System Sciences","day":"03","year":"2021","isi":1,"status":"public","type":"journal_article","article_type":"original","_id":"9239","department":[{"_id":"ToHe"}],"date_updated":"2023-08-07T14:08:34Z","intvolume":" 119","month":"03","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.1905.03835","open_access":"1"}],"scopus_import":"1","oa_version":"Preprint","abstract":[{"text":"A graph game proceeds as follows: two players move a token through a graph to produce a finite or infinite path, which determines the payoff of the game. We study bidding games in which in each turn, an auction determines which player moves the token. Bidding games were largely studied in combination with two variants of first-price auctions called “Richman” and “poorman” bidding. We study taxman bidding, which span the spectrum between the two. The game is parameterized by a constant : portion τ of the winning bid is paid to the other player, and portion to the bank. While finite-duration (reachability) taxman games have been studied before, we present, for the first time, results on infinite-duration taxman games: we unify, generalize, and simplify previous equivalences between bidding games and a class of stochastic games called random-turn games.","lang":"eng"}],"related_material":{"record":[{"relation":"earlier_version","id":"6884","status":"public"}]},"issue":"8","volume":119,"language":[{"iso":"eng"}],"publication_status":"published","publication_identifier":{"issn":["0022-0000"],"eissn":["1090-2724"]}},{"publication_status":"published","file":[{"file_name":"qam.pdf","date_created":"2021-06-16T08:23:54Z","creator":"esarac","file_size":641990,"date_updated":"2021-06-16T08:23:54Z","success":1,"checksum":"6e4cba3f72775f479c5b1b75d1a4a0c4","file_id":"9557","relation":"main_file","access_level":"open_access","content_type":"application/pdf"}],"language":[{"iso":"eng"}],"scopus_import":"1","month":"06","abstract":[{"text":"In runtime verification, a monitor watches a trace of a system and, if possible, decides after observing each finite prefix whether or not the unknown infinite trace satisfies a given specification. We generalize the theory of runtime verification to monitors that attempt to estimate numerical values of quantitative trace properties (instead of attempting to conclude boolean values of trace specifications), such as maximal or average response time along a trace. Quantitative monitors are approximate: with every finite prefix, they can improve their estimate of the infinite trace's unknown property value. Consequently, quantitative monitors can be compared with regard to a precision-cost trade-off: better approximations of the property value require more monitor resources, such as states (in the case of finite-state monitors) or registers, and additional resources yield better approximations. We introduce a formal framework for quantitative and approximate monitoring, show how it conservatively generalizes the classical boolean setting for monitoring, and give several precision-cost trade-offs for monitors. For example, we prove that there are quantitative properties for which every additional register improves monitoring precision.","lang":"eng"}],"oa_version":"Published Version","file_date_updated":"2021-06-16T08:23:54Z","department":[{"_id":"GradSch"},{"_id":"ToHe"}],"date_updated":"2023-08-08T13:52:56Z","ddc":["000"],"type":"conference","conference":{"location":"Online","end_date":"2021-07-02","start_date":"2021-06-29","name":"LICS: Symposium on Logic in Computer Science"},"status":"public","_id":"9356","doi":"10.1109/LICS52264.2021.9470547","date_published":"2021-06-29T00:00:00Z","date_created":"2021-04-30T17:30:47Z","has_accepted_license":"1","isi":1,"year":"2021","day":"29","publication":"Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science","quality_controlled":"1","publisher":"Institute of Electrical and Electronics Engineers","oa":1,"acknowledgement":"We thank the anonymous reviewers for their helpful comments. This research was supported in part by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award).","author":[{"first_name":"Thomas A","id":"40876CD8-F248-11E8-B48F-1D18A9856A87","last_name":"Henzinger","full_name":"Henzinger, Thomas A","orcid":"0000-0002-2985-7724"},{"last_name":"Sarac","full_name":"Sarac, Naci E","first_name":"Naci E","id":"8C6B42F8-C8E6-11E9-A03A-F2DCE5697425"}],"article_processing_charge":"No","external_id":{"arxiv":["2105.08353"],"isi":["000947350400021"]},"title":"Quantitative and approximate monitoring","citation":{"ieee":"T. A. Henzinger and N. E. Sarac, “Quantitative and approximate monitoring,” in Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science, Online, 2021.","short":"T.A. Henzinger, N.E. Sarac, in:, Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science, Institute of Electrical and Electronics Engineers, 2021.","ama":"Henzinger TA, Sarac NE. Quantitative and approximate monitoring. In: Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science. Institute of Electrical and Electronics Engineers; 2021. doi:10.1109/LICS52264.2021.9470547","apa":"Henzinger, T. A., & Sarac, N. E. (2021). Quantitative and approximate monitoring. In Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science. Online: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/LICS52264.2021.9470547","mla":"Henzinger, Thomas A., and Naci E. Sarac. “Quantitative and Approximate Monitoring.” Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science, 9470547, Institute of Electrical and Electronics Engineers, 2021, doi:10.1109/LICS52264.2021.9470547.","ista":"Henzinger TA, Sarac NE. 2021. Quantitative and approximate monitoring. Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science. LICS: Symposium on Logic in Computer Science, 9470547.","chicago":"Henzinger, Thomas A, and Naci E Sarac. “Quantitative and Approximate Monitoring.” In Proceedings of the 36th Annual ACM/IEEE Symposium on Logic in Computer Science. Institute of Electrical and Electronics Engineers, 2021. https://doi.org/10.1109/LICS52264.2021.9470547."},"user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","project":[{"grant_number":"Z211","name":"The Wittgenstein Prize","_id":"25F42A32-B435-11E9-9278-68D0E5697425","call_identifier":"FWF"}],"article_number":"9470547"}]