--- _id: '6985' abstract: - lang: eng text: In this paper, we introduce a novel method to interpret recurrent neural networks (RNNs), particularly long short-term memory networks (LSTMs) at the cellular level. We propose a systematic pipeline for interpreting individual hidden state dynamics within the network using response characterization methods. The ranked contribution of individual cells to the network's output is computed by analyzing a set of interpretable metrics of their decoupled step and sinusoidal responses. As a result, our method is able to uniquely identify neurons with insightful dynamics, quantify relationships between dynamical properties and test accuracy through ablation analysis, and interpret the impact of network capacity on a network's dynamical distribution. Finally, we demonstrate the generalizability and scalability of our method by evaluating a series of different benchmark sequential datasets. article_number: '8851954' author: - first_name: Ramin full_name: Hasani, Ramin last_name: Hasani - first_name: Alexander full_name: Amini, Alexander last_name: Amini - first_name: Mathias full_name: Lechner, Mathias id: 3DC22916-F248-11E8-B48F-1D18A9856A87 last_name: Lechner - first_name: Felix full_name: Naser, Felix last_name: Naser - first_name: Radu full_name: Grosu, Radu last_name: Grosu - first_name: Daniela full_name: Rus, Daniela last_name: Rus citation: ama: 'Hasani R, Amini A, Lechner M, Naser F, Grosu R, Rus D. Response characterization for auditing cell dynamics in long short-term memory networks. In: Proceedings of the International Joint Conference on Neural Networks. IEEE; 2019. doi:10.1109/ijcnn.2019.8851954' apa: 'Hasani, R., Amini, A., Lechner, M., Naser, F., Grosu, R., & Rus, D. (2019). Response characterization for auditing cell dynamics in long short-term memory networks. In Proceedings of the International Joint Conference on Neural Networks. Budapest, Hungary: IEEE. https://doi.org/10.1109/ijcnn.2019.8851954' chicago: Hasani, Ramin, Alexander Amini, Mathias Lechner, Felix Naser, Radu Grosu, and Daniela Rus. “Response Characterization for Auditing Cell Dynamics in Long Short-Term Memory Networks.” In Proceedings of the International Joint Conference on Neural Networks. IEEE, 2019. https://doi.org/10.1109/ijcnn.2019.8851954. ieee: R. Hasani, A. Amini, M. Lechner, F. Naser, R. Grosu, and D. Rus, “Response characterization for auditing cell dynamics in long short-term memory networks,” in Proceedings of the International Joint Conference on Neural Networks, Budapest, Hungary, 2019. ista: 'Hasani R, Amini A, Lechner M, Naser F, Grosu R, Rus D. 2019. Response characterization for auditing cell dynamics in long short-term memory networks. Proceedings of the International Joint Conference on Neural Networks. IJCNN: International Joint Conference on Neural Networks, 8851954.' mla: Hasani, Ramin, et al. “Response Characterization for Auditing Cell Dynamics in Long Short-Term Memory Networks.” Proceedings of the International Joint Conference on Neural Networks, 8851954, IEEE, 2019, doi:10.1109/ijcnn.2019.8851954. short: R. Hasani, A. Amini, M. Lechner, F. Naser, R. Grosu, D. Rus, in:, Proceedings of the International Joint Conference on Neural Networks, IEEE, 2019. conference: end_date: 2019-07-19 location: Budapest, Hungary name: 'IJCNN: International Joint Conference on Neural Networks' start_date: 2019-07-14 date_created: 2019-11-04T15:59:58Z date_published: 2019-09-30T00:00:00Z date_updated: 2021-01-12T08:11:19Z day: '30' department: - _id: ToHe doi: 10.1109/ijcnn.2019.8851954 external_id: arxiv: - '1809.03864' language: - iso: eng main_file_link: - open_access: '1' url: https://arxiv.org/abs/1809.03864 month: '09' oa: 1 oa_version: Preprint publication: Proceedings of the International Joint Conference on Neural Networks publication_identifier: isbn: - '9781728119854' publication_status: published publisher: IEEE quality_controlled: '1' scopus_import: 1 status: public title: Response characterization for auditing cell dynamics in long short-term memory networks type: conference user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87 year: '2019' ... --- _id: '7007' abstract: - lang: eng text: 'We consider the primitive relay channel, where the source sends a message to the relay and to the destination, and the relay helps the communication by transmitting an additional message to the destination via a separate channel. Two well-known coding techniques have been introduced for this setting: decode-and-forward and compress-and-forward. In decode-and-forward, the relay completely decodes the message and sends some information to the destination; in compress-and-forward, the relay does not decode, and it sends a compressed version of the received signal to the destination using Wyner–Ziv coding. In this paper, we present a novel coding paradigm that provides an improved achievable rate for the primitive relay channel. The idea is to combine compress-and-forward and decode-and-forward via a chaining construction. We transmit over pairs of blocks: in the first block, we use compress-and-forward; and, in the second block, we use decode-and-forward. More specifically, in the first block, the relay does not decode, it compresses the received signal via Wyner–Ziv, and it sends only part of the compression to the destination. In the second block, the relay completely decodes the message, it sends some information to the destination, and it also sends the remaining part of the compression coming from the first block. By doing so, we are able to strictly outperform both compress-and-forward and decode-and-forward. Note that the proposed coding scheme can be implemented with polar codes. As such, it has the typical attractive properties of polar coding schemes, namely, quasi-linear encoding and decoding complexity, and error probability that decays at super-polynomial speed. As a running example, we take into account the special case of the erasure relay channel, and we provide a comparison between the rates achievable by our proposed scheme and the existing upper and lower bounds.' article_number: '218' article_type: original author: - first_name: Marco full_name: Mondelli, Marco id: 27EB676C-8706-11E9-9510-7717E6697425 last_name: Mondelli orcid: 0000-0002-3242-7020 - first_name: S. Hamed full_name: Hassani, S. Hamed last_name: Hassani - first_name: Rüdiger full_name: Urbanke, Rüdiger last_name: Urbanke citation: ama: Mondelli M, Hassani SH, Urbanke R. A new coding paradigm for the primitive relay channel. Algorithms. 2019;12(10). doi:10.3390/a12100218 apa: Mondelli, M., Hassani, S. H., & Urbanke, R. (2019). A new coding paradigm for the primitive relay channel. Algorithms. MDPI. https://doi.org/10.3390/a12100218 chicago: Mondelli, Marco, S. Hamed Hassani, and Rüdiger Urbanke. “A New Coding Paradigm for the Primitive Relay Channel.” Algorithms. MDPI, 2019. https://doi.org/10.3390/a12100218. ieee: M. Mondelli, S. H. Hassani, and R. Urbanke, “A new coding paradigm for the primitive relay channel,” Algorithms, vol. 12, no. 10. MDPI, 2019. ista: Mondelli M, Hassani SH, Urbanke R. 2019. A new coding paradigm for the primitive relay channel. Algorithms. 12(10), 218. mla: Mondelli, Marco, et al. “A New Coding Paradigm for the Primitive Relay Channel.” Algorithms, vol. 12, no. 10, 218, MDPI, 2019, doi:10.3390/a12100218. short: M. Mondelli, S.H. Hassani, R. Urbanke, Algorithms 12 (2019). date_created: 2019-11-12T14:46:19Z date_published: 2019-10-18T00:00:00Z date_updated: 2023-02-23T12:49:28Z day: '18' ddc: - '510' department: - _id: MaMo doi: 10.3390/a12100218 external_id: arxiv: - '1801.03153' file: - access_level: open_access checksum: 267756d8f9db572f496cd1663c89d59a content_type: application/pdf creator: dernst date_created: 2019-11-12T14:48:45Z date_updated: 2020-07-14T12:47:47Z file_id: '7008' file_name: 2019_Algorithms_Mondelli.pdf file_size: 696791 relation: main_file file_date_updated: 2020-07-14T12:47:47Z has_accepted_license: '1' intvolume: ' 12' issue: '10' language: - iso: eng license: https://creativecommons.org/licenses/by/4.0/ month: '10' oa: 1 oa_version: Published Version publication: Algorithms publication_identifier: issn: - 1999-4893 publication_status: published publisher: MDPI quality_controlled: '1' related_material: record: - id: '6675' relation: earlier_version status: public scopus_import: 1 status: public title: A new coding paradigm for the primitive relay channel tmp: image: /images/cc_by.png legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0) short: CC BY (4.0) type: journal_article user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87 volume: 12 year: '2019' ... --- _id: '7035' abstract: - lang: eng text: 'The aim of this short note is to expound one particular issue that was discussed during the talk [10] given at the symposium ”Researches on isometries as preserver problems and related topics” at Kyoto RIMS. That is, the role of Dirac masses by describing the isometry group of various metric spaces of probability measures. This article is of survey character, and it does not contain any essentially new results.From an isometric point of view, in some cases, metric spaces of measures are similar to C(K)-type function spaces. Similarity means here that their isometries are driven by some nice transformations of the underlying space. Of course, it depends on the particular choice of the metric how nice these transformations should be. Sometimes, as we will see, being a homeomorphism is enough to generate an isometry. But sometimes we need more: the transformation must preserve the underlying distance as well. Statements claiming that isometries in questions are necessarily induced by homeomorphisms are called Banach-Stone-type results, while results asserting that the underlying transformation is necessarily an isometry are termed as isometric rigidity results.As Dirac masses can be considered as building bricks of the set of all Borel measures, a natural question arises:Is it enough to understand how an isometry acts on the set of Dirac masses? Does this action extend uniquely to all measures?In what follows, we will thoroughly investigate this question.' article_processing_charge: No author: - first_name: Gyorgy Pal full_name: Geher, Gyorgy Pal last_name: Geher - first_name: Tamas full_name: Titkos, Tamas last_name: Titkos - first_name: Daniel full_name: Virosztek, Daniel id: 48DB45DA-F248-11E8-B48F-1D18A9856A87 last_name: Virosztek orcid: 0000-0003-1109-5511 citation: ama: 'Geher GP, Titkos T, Virosztek D. Dirac masses and isometric rigidity. In: Kyoto RIMS Kôkyûroku. Vol 2125. Research Institute for Mathematical Sciences, Kyoto University; 2019:34-41.' apa: 'Geher, G. P., Titkos, T., & Virosztek, D. (2019). Dirac masses and isometric rigidity. In Kyoto RIMS Kôkyûroku (Vol. 2125, pp. 34–41). Kyoto, Japan: Research Institute for Mathematical Sciences, Kyoto University.' chicago: Geher, Gyorgy Pal, Tamas Titkos, and Daniel Virosztek. “Dirac Masses and Isometric Rigidity.” In Kyoto RIMS Kôkyûroku, 2125:34–41. Research Institute for Mathematical Sciences, Kyoto University, 2019. ieee: G. P. Geher, T. Titkos, and D. Virosztek, “Dirac masses and isometric rigidity,” in Kyoto RIMS Kôkyûroku, Kyoto, Japan, 2019, vol. 2125, pp. 34–41. ista: Geher GP, Titkos T, Virosztek D. 2019. Dirac masses and isometric rigidity. Kyoto RIMS Kôkyûroku. Research on isometries as preserver problems and related topics vol. 2125, 34–41. mla: Geher, Gyorgy Pal, et al. “Dirac Masses and Isometric Rigidity.” Kyoto RIMS Kôkyûroku, vol. 2125, Research Institute for Mathematical Sciences, Kyoto University, 2019, pp. 34–41. short: G.P. Geher, T. Titkos, D. Virosztek, in:, Kyoto RIMS Kôkyûroku, Research Institute for Mathematical Sciences, Kyoto University, 2019, pp. 34–41. conference: end_date: 2019-01-30 location: Kyoto, Japan name: Research on isometries as preserver problems and related topics start_date: 2019-01-28 date_created: 2019-11-18T15:39:53Z date_published: 2019-01-30T00:00:00Z date_updated: 2021-01-12T08:11:33Z day: '30' department: - _id: LaEr intvolume: ' 2125' language: - iso: eng main_file_link: - open_access: '1' url: http://www.kurims.kyoto-u.ac.jp/~kyodo/kokyuroku/contents/2125.html month: '01' oa: 1 oa_version: Submitted Version page: 34-41 publication: Kyoto RIMS Kôkyûroku publication_status: published publisher: Research Institute for Mathematical Sciences, Kyoto University quality_controlled: '1' status: public title: Dirac masses and isometric rigidity type: conference user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87 volume: 2125 year: '2019' ... --- _id: '7171' abstract: - lang: ger text: "Wissen Sie, was sich hinter künstlicher Intelligenz und maschinellem Lernen verbirgt? \r\nDieses Sachbuch erklärt Ihnen leicht verständlich und ohne komplizierte Formeln die grundlegenden Methoden und Vorgehensweisen des maschinellen Lernens. Mathematisches Vorwissen ist dafür nicht nötig. Kurzweilig und informativ illustriert Lisa, die Protagonistin des Buches, diese anhand von Alltagssituationen. \r\nEin Buch für alle, die in Diskussionen über Chancen und Risiken der aktuellen Entwicklung der künstlichen Intelligenz und des maschinellen Lernens mit Faktenwissen punkten möchten. Auch für Schülerinnen und Schüler geeignet!" article_processing_charge: No citation: ama: 'Kersting K, Lampert C, Rothkopf C, eds. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt. 1st ed. Wiesbaden: Springer Nature; 2019. doi:10.1007/978-3-658-26763-6' apa: 'Kersting, K., Lampert, C., & Rothkopf, C. (Eds.). (2019). Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt (1st ed.). Wiesbaden: Springer Nature. https://doi.org/10.1007/978-3-658-26763-6' chicago: 'Kersting, Kristian, Christoph Lampert, and Constantin Rothkopf, eds. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt. 1st ed. Wiesbaden: Springer Nature, 2019. https://doi.org/10.1007/978-3-658-26763-6.' ieee: 'K. Kersting, C. Lampert, and C. Rothkopf, Eds., Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt, 1st ed. Wiesbaden: Springer Nature, 2019.' ista: 'Kersting K, Lampert C, Rothkopf C eds. 2019. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt 1st ed., Wiesbaden: Springer Nature, XIV, 245p.' mla: 'Kersting, Kristian, et al., editors. Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt. 1st ed., Springer Nature, 2019, doi:10.1007/978-3-658-26763-6.' short: 'K. Kersting, C. Lampert, C. Rothkopf, eds., Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt, 1st ed., Springer Nature, Wiesbaden, 2019.' date_created: 2019-12-11T14:15:56Z date_published: 2019-10-30T00:00:00Z date_updated: 2021-12-22T14:40:58Z day: '30' department: - _id: ChLa doi: 10.1007/978-3-658-26763-6 edition: '1' editor: - first_name: Kristian full_name: Kersting, Kristian last_name: Kersting - first_name: Christoph full_name: Lampert, Christoph id: 40C20FD2-F248-11E8-B48F-1D18A9856A87 last_name: Lampert orcid: 0000-0001-8622-7887 - first_name: Constantin full_name: Rothkopf, Constantin last_name: Rothkopf language: - iso: ger month: '10' oa_version: None page: XIV, 245 place: Wiesbaden publication_identifier: eisbn: - 978-3-658-26763-6 isbn: - 978-3-658-26762-9 publication_status: published publisher: Springer Nature quality_controlled: '1' related_material: link: - description: News on IST Website relation: press_release url: https://ist.ac.at/en/news/book-release-how-machines-learn/ status: public title: 'Wie Maschinen Lernen: Künstliche Intelligenz Verständlich Erklärt' type: book_editor user_id: 8b945eb4-e2f2-11eb-945a-df72226e66a9 year: '2019' ... --- _id: '7401' abstract: - lang: eng text: 'The genus g(G) of a graph G is the minimum g such that G has an embedding on the orientable surface M_g of genus g. A drawing of a graph on a surface is independently even if every pair of nonadjacent edges in the drawing crosses an even number of times. The Z_2-genus of a graph G, denoted by g_0(G), is the minimum g such that G has an independently even drawing on M_g. By a result of Battle, Harary, Kodama and Youngs from 1962, the graph genus is additive over 2-connected blocks. In 2013, Schaefer and Stefankovic proved that the Z_2-genus of a graph is additive over 2-connected blocks as well, and asked whether this result can be extended to so-called 2-amalgamations, as an analogue of results by Decker, Glover, Huneke, and Stahl for the genus. We give the following partial answer. If G=G_1 cup G_2, G_1 and G_2 intersect in two vertices u and v, and G-u-v has k connected components (among which we count the edge uv if present), then |g_0(G)-(g_0(G_1)+g_0(G_2))|<=k+1. For complete bipartite graphs K_{m,n}, with n >= m >= 3, we prove that g_0(K_{m,n})/g(K_{m,n})=1-O(1/n). Similar results are proved also for the Euler Z_2-genus. We express the Z_2-genus of a graph using the minimum rank of partial symmetric matrices over Z_2; a problem that might be of independent interest. ' alternative_title: - LIPIcs article_number: '39' article_processing_charge: No author: - first_name: Radoslav full_name: Fulek, Radoslav id: 39F3FFE4-F248-11E8-B48F-1D18A9856A87 last_name: Fulek orcid: 0000-0001-8485-1774 - first_name: Jan full_name: Kyncl, Jan last_name: Kyncl citation: ama: 'Fulek R, Kyncl J. Z_2-Genus of graphs and minimum rank of partial symmetric matrices. In: 35th International Symposium on Computational Geometry (SoCG 2019). Vol 129. Schloss Dagstuhl - Leibniz-Zentrum für Informatik; 2019. doi:10.4230/LIPICS.SOCG.2019.39' apa: 'Fulek, R., & Kyncl, J. (2019). Z_2-Genus of graphs and minimum rank of partial symmetric matrices. In 35th International Symposium on Computational Geometry (SoCG 2019) (Vol. 129). Portland, OR, United States: Schloss Dagstuhl - Leibniz-Zentrum für Informatik. https://doi.org/10.4230/LIPICS.SOCG.2019.39' chicago: Fulek, Radoslav, and Jan Kyncl. “Z_2-Genus of Graphs and Minimum Rank of Partial Symmetric Matrices.” In 35th International Symposium on Computational Geometry (SoCG 2019), Vol. 129. Schloss Dagstuhl - Leibniz-Zentrum für Informatik, 2019. https://doi.org/10.4230/LIPICS.SOCG.2019.39. ieee: R. Fulek and J. Kyncl, “Z_2-Genus of graphs and minimum rank of partial symmetric matrices,” in 35th International Symposium on Computational Geometry (SoCG 2019), Portland, OR, United States, 2019, vol. 129. ista: 'Fulek R, Kyncl J. 2019. Z_2-Genus of graphs and minimum rank of partial symmetric matrices. 35th International Symposium on Computational Geometry (SoCG 2019). SoCG: Symposium on Computational Geometry, LIPIcs, vol. 129, 39.' mla: Fulek, Radoslav, and Jan Kyncl. “Z_2-Genus of Graphs and Minimum Rank of Partial Symmetric Matrices.” 35th International Symposium on Computational Geometry (SoCG 2019), vol. 129, 39, Schloss Dagstuhl - Leibniz-Zentrum für Informatik, 2019, doi:10.4230/LIPICS.SOCG.2019.39. short: R. Fulek, J. Kyncl, in:, 35th International Symposium on Computational Geometry (SoCG 2019), Schloss Dagstuhl - Leibniz-Zentrum für Informatik, 2019. conference: end_date: 2019-06-21 location: Portland, OR, United States name: 'SoCG: Symposium on Computational Geometry' start_date: 2019-06-18 date_created: 2020-01-29T16:17:05Z date_published: 2019-06-01T00:00:00Z date_updated: 2021-01-12T08:13:24Z day: '01' ddc: - '000' department: - _id: UlWa doi: 10.4230/LIPICS.SOCG.2019.39 external_id: arxiv: - '1903.08637' file: - access_level: open_access checksum: aac37b09118cc0ab58cf77129e691f8c content_type: application/pdf creator: dernst date_created: 2020-02-04T09:14:31Z date_updated: 2020-07-14T12:47:57Z file_id: '7445' file_name: 2019_LIPIcs_Fulek.pdf file_size: 628347 relation: main_file file_date_updated: 2020-07-14T12:47:57Z has_accepted_license: '1' intvolume: ' 129' language: - iso: eng month: '06' oa: 1 oa_version: Published Version project: - _id: 261FA626-B435-11E9-9278-68D0E5697425 call_identifier: FWF grant_number: M02281 name: Eliminating intersections in drawings of graphs publication: 35th International Symposium on Computational Geometry (SoCG 2019) publication_identifier: isbn: - 978-3-95977-104-7 issn: - 1868-8969 publication_status: published publisher: Schloss Dagstuhl - Leibniz-Zentrum für Informatik quality_controlled: '1' scopus_import: 1 status: public title: Z_2-Genus of graphs and minimum rank of partial symmetric matrices tmp: image: /images/cc_by.png legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0) short: CC BY (4.0) type: conference user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87 volume: 129 year: '2019' ...