[{"OA_type":"green","citation":{"short":"A. Casares, M. Pilipczuk, M. Pilipczuk, U.S. Souza, K.S. Thejaswini, in:, 2024 Symposium on Simplicity in Algorithms, Society for Industrial and Applied Mathematics, 2024, pp. 160–167.","apa":"Casares, A., Pilipczuk, M., Pilipczuk, M., Souza, U. S., &#38; Thejaswini, K. S. (2024). Simple and tight complexity lower bounds for solving Rabin games. In <i>2024 Symposium on Simplicity in Algorithms</i> (pp. 160–167). Alexandria, VA, United States: Society for Industrial and Applied Mathematics. <a href=\"https://doi.org/10.1137/1.9781611977936.16\">https://doi.org/10.1137/1.9781611977936.16</a>","ama":"Casares A, Pilipczuk M, Pilipczuk M, Souza US, Thejaswini KS. Simple and tight complexity lower bounds for solving Rabin games. In: <i>2024 Symposium on Simplicity in Algorithms</i>. Society for Industrial and Applied Mathematics; 2024:160-167. doi:<a href=\"https://doi.org/10.1137/1.9781611977936.16\">10.1137/1.9781611977936.16</a>","ista":"Casares A, Pilipczuk M, Pilipczuk M, Souza US, Thejaswini KS. 2024. Simple and tight complexity lower bounds for solving Rabin games. 2024 Symposium on Simplicity in Algorithms. SOSA: Symposium on Simplicity in Algorithms, 160–167.","mla":"Casares, Antonio, et al. “Simple and Tight Complexity Lower Bounds for Solving Rabin Games.” <i>2024 Symposium on Simplicity in Algorithms</i>, Society for Industrial and Applied Mathematics, 2024, pp. 160–67, doi:<a href=\"https://doi.org/10.1137/1.9781611977936.16\">10.1137/1.9781611977936.16</a>.","ieee":"A. Casares, M. Pilipczuk, M. Pilipczuk, U. S. Souza, and K. S. Thejaswini, “Simple and tight complexity lower bounds for solving Rabin games,” in <i>2024 Symposium on Simplicity in Algorithms</i>, Alexandria, VA, United States, 2024, pp. 160–167.","chicago":"Casares, Antonio, Marcin Pilipczuk, Michał Pilipczuk, Uéverton S. Souza, and K. S. Thejaswini. “Simple and Tight Complexity Lower Bounds for Solving Rabin Games.” In <i>2024 Symposium on Simplicity in Algorithms</i>, 160–67. Society for Industrial and Applied Mathematics, 2024. <a href=\"https://doi.org/10.1137/1.9781611977936.16\">https://doi.org/10.1137/1.9781611977936.16</a>."},"ec_funded":1,"OA_place":"repository","oa_version":"Preprint","page":"160-167","language":[{"iso":"eng"}],"month":"01","conference":{"location":"Alexandria, VA, United States","start_date":"2024-01-08","end_date":"2024-01-10","name":"SOSA: Symposium on Simplicity in Algorithms"},"department":[{"_id":"ToHe"}],"author":[{"first_name":"Antonio","last_name":"Casares","full_name":"Casares, Antonio"},{"last_name":"Pilipczuk","full_name":"Pilipczuk, Marcin","first_name":"Marcin"},{"first_name":"Michał","last_name":"Pilipczuk","full_name":"Pilipczuk, Michał"},{"last_name":"Souza","full_name":"Souza, Uéverton S.","first_name":"Uéverton S."},{"id":"3807fb92-fdc1-11ee-bb4a-b4d8a431c753","first_name":"K. S.","full_name":"Thejaswini, K. S.","last_name":"Thejaswini"}],"publication_status":"published","acknowledgement":"This work is a part of projects CUTACOMBS (Ma. Pilipczuk), BOBR (Mi. Pilipczuk), and VAMOS (K. S. Thejaswini) that have received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme, grant agreements No 714704, 948057, and 101020093, respectively. Ma. Pilipczuk is also partially supported by Polish National Science Centre SONATA BIS-12 grant number 2022/46/E/ST6/00143.","scopus_import":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2310.20433"}],"publication_identifier":{"isbn":["9781611977936"]},"title":"Simple and tight complexity lower bounds for solving Rabin games","quality_controlled":"1","abstract":[{"lang":"eng","text":"We give a simple proof that assuming the Exponential Time Hypothesis (ETH), determining the winner of a Rabin game cannot be done in time 2o(k log k) · nO(1), where k is the number of pairs of vertex subsets involved in the winning condition and n is the vertex count of the game graph. While this result follows from the lower bounds provided by Calude et al [SIAM J. Comp. 2022], our reduction is considerably simpler and arguably provides more insight into the complexity of the problem. In fact, the analogous lower bounds discussed by Calude et al, for solving Muller games and multidimensional parity games, follow as simple corollaries of our approach. Our reduction also highlights the usefulness of a certain pivot problem — Permutation SAT — which may be of independent interest."}],"publisher":"Society for Industrial and Applied Mathematics","oa":1,"date_updated":"2025-04-14T07:55:54Z","status":"public","doi":"10.1137/1.9781611977936.16","project":[{"_id":"62781420-2b32-11ec-9570-8d9b63373d4d","grant_number":"101020093","name":"Vigilant Algorithmic Monitoring of Software","call_identifier":"H2020"}],"arxiv":1,"year":"2024","_id":"18955","external_id":{"arxiv":["2310.20433"]},"date_published":"2024-01-01T00:00:00Z","date_created":"2025-01-29T11:55:50Z","publication":"2024 Symposium on Simplicity in Algorithms","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","day":"01","article_processing_charge":"No"},{"publication_identifier":{"eisbn":["9798350349399"],"eissn":["2381-8549"]},"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2303.06439","open_access":"1"}],"publisher":"IEEE","title":"Decompl: Decompositional learning with attention pooling for group activity recognition from a single volleyball image","abstract":[{"lang":"eng","text":"Group Activity Recognition (GAR) aims to detect the activity performed by multiple actors in a scene. Prior works model the spatio-temporal features based on the RGB, optical flow or keypoint data types. On the contrary, our hypothesis is that by only using the RGB data without temporality, the performance can be maintained with a negligible loss in accuracy. To that end, we propose a novel GAR technique for volleyball videos, DECOMPL, which consists of two complementary branches. In the visual branch, it extracts the features using attention pooling. In the coordinate branch, it considers the configuration of the players and extracts the spatial information from the box coordinates. Moreover, we analyzed the Volleyball dataset that the recent literature is mostly based on, and systematically reannotated it to emphasize the group concept. Experimental results demonstrated the effectiveness of the proposed model DECOMPL, which delivered the best/second best GAR performance with the reannotations/original annotations among the comparable state-of-the-art methods. Code and new annotations are available at GitHub: https://github.com/berkerdemirel/decompl"}],"quality_controlled":"1","oa":1,"arxiv":1,"doi":"10.1109/icip51287.2024.10647499","status":"public","date_updated":"2025-09-09T12:13:12Z","year":"2024","date_created":"2025-01-29T12:22:24Z","date_published":"2024-11-01T00:00:00Z","_id":"18956","external_id":{"arxiv":["2303.06439"],"isi":["001442947000143"]},"user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","type":"conference","publication":"2024 IEEE International Conference on Image Processing","article_processing_charge":"No","day":"01","corr_author":"1","citation":{"ama":"Demirel B, Ozkan H. Decompl: Decompositional learning with attention pooling for group activity recognition from a single volleyball image. In: <i>2024 IEEE International Conference on Image Processing</i>. IEEE; 2024:977-983. doi:<a href=\"https://doi.org/10.1109/icip51287.2024.10647499\">10.1109/icip51287.2024.10647499</a>","apa":"Demirel, B., &#38; Ozkan, H. (2024). Decompl: Decompositional learning with attention pooling for group activity recognition from a single volleyball image. In <i>2024 IEEE International Conference on Image Processing</i> (pp. 977–983). Abu Dhabi, United Arab Emirates: IEEE. <a href=\"https://doi.org/10.1109/icip51287.2024.10647499\">https://doi.org/10.1109/icip51287.2024.10647499</a>","short":"B. Demirel, H. Ozkan, in:, 2024 IEEE International Conference on Image Processing, IEEE, 2024, pp. 977–983.","ieee":"B. Demirel and H. Ozkan, “Decompl: Decompositional learning with attention pooling for group activity recognition from a single volleyball image,” in <i>2024 IEEE International Conference on Image Processing</i>, Abu Dhabi, United Arab Emirates, 2024, pp. 977–983.","chicago":"Demirel, Berker, and Huseyin Ozkan. “Decompl: Decompositional Learning with Attention Pooling for Group Activity Recognition from a Single Volleyball Image.” In <i>2024 IEEE International Conference on Image Processing</i>, 977–83. IEEE, 2024. <a href=\"https://doi.org/10.1109/icip51287.2024.10647499\">https://doi.org/10.1109/icip51287.2024.10647499</a>.","mla":"Demirel, Berker, and Huseyin Ozkan. “Decompl: Decompositional Learning with Attention Pooling for Group Activity Recognition from a Single Volleyball Image.” <i>2024 IEEE International Conference on Image Processing</i>, IEEE, 2024, pp. 977–83, doi:<a href=\"https://doi.org/10.1109/icip51287.2024.10647499\">10.1109/icip51287.2024.10647499</a>.","ista":"Demirel B, Ozkan H. 2024. Decompl: Decompositional learning with attention pooling for group activity recognition from a single volleyball image. 2024 IEEE International Conference on Image Processing. ICIP: International Conference on Image Processing, 977–983."},"OA_type":"green","related_material":{"link":[{"url":"https://github.com/berkerdemirel/decompl","relation":"software"}]},"OA_place":"repository","language":[{"iso":"eng"}],"page":"977-983","oa_version":"Preprint","publication_status":"published","isi":1,"department":[{"_id":"FrLo"}],"conference":{"end_date":"2024-10-30","location":"Abu Dhabi, United Arab Emirates","start_date":"2024-10-27","name":"ICIP: International Conference on Image Processing"},"author":[{"id":"8b4bc47f-3200-11ee-973b-8f0e7be21a9f","first_name":"Berker","last_name":"Demirel","full_name":"Demirel, Berker"},{"last_name":"Ozkan","full_name":"Ozkan, Huseyin","first_name":"Huseyin"}],"month":"11"},{"acknowledgement":"This work is funded by MystenLabs. We thank the Mysten Labs Engineering teams for valuable feedback broadly, and specifically Dmitry Perelman and Todd Fiala for managing the implementation effort. A number of folks contributed to specific aspects of the implementation of Sui Lutris (amongst many other contributions to the overall blockchain): Francois Garillot, Laura Makdah, Mingwei Tian, Andrew Schran, Sadhan Sood and William Smith implemented and optimized aspects of both Sui Lutris and Narwhal / Bullshark consensus; Alonso de Gortari oversaw the cryptoeconomics of the blockchain, and Emma Zhong, Ade Adepoju, Tim Zakia and Dario Russi designed and implemented staking and gas mechanisms. Adam Welc designed several Move tools and provided great feedback on the manuscript. We also extend our thanks to Patrick Kuo, Ge Gao, Chris Li, and Arun Koshy for their work on the Sui Lutris SDK, clients, and RPC layer; Kostas Chalkias, Jonas Lindstrøm, and Joy Wang built cryptographic components.","month":"12","publication_status":"published","isi":1,"author":[{"first_name":"Sam","full_name":"Blackshear, Sam","last_name":"Blackshear"},{"last_name":"Chursin","full_name":"Chursin, Andrey","first_name":"Andrey"},{"last_name":"Danezis","full_name":"Danezis, George","first_name":"George"},{"full_name":"Kichidis, Anastasios","last_name":"Kichidis","first_name":"Anastasios"},{"first_name":"Eleftherios","id":"f5983044-d7ef-11ea-ac6d-fd1430a26d30","full_name":"Kokoris Kogias, Eleftherios","last_name":"Kokoris Kogias"},{"first_name":"Xun","full_name":"Li, Xun","last_name":"Li"},{"last_name":"Logan","full_name":"Logan, Mark","first_name":"Mark"},{"first_name":"Ashok","last_name":"Menon","full_name":"Menon, Ashok"},{"full_name":"Nowacki, Todd","last_name":"Nowacki","first_name":"Todd"},{"full_name":"Sonnino, Alberto","last_name":"Sonnino","first_name":"Alberto"},{"first_name":"Brandon","last_name":"Williams","full_name":"Williams, Brandon"},{"first_name":"Lu","last_name":"Zhang","full_name":"Zhang, Lu"}],"department":[{"_id":"ElKo"}],"conference":{"name":"CCS: Conference on Computer and Communications Security","start_date":"2024-10-14","location":"Salt Lake City, UT, United States","end_date":"2024-10-18"},"page":"2606-2620","oa_version":"Preprint","language":[{"iso":"eng"}],"OA_place":"repository","OA_type":"hybrid","citation":{"short":"S. Blackshear, A. Chursin, G. Danezis, A. Kichidis, E. Kokoris Kogias, X. Li, M. Logan, A. Menon, T. Nowacki, A. Sonnino, B. Williams, L. Zhang, in:, Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security, ACM, 2024, pp. 2606–2620.","ama":"Blackshear S, Chursin A, Danezis G, et al. Sui Lutris: A blockchain combining broadcast and consensus. In: <i>Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security</i>. ACM; 2024:2606-2620. doi:<a href=\"https://doi.org/10.1145/3658644.3670286\">10.1145/3658644.3670286</a>","apa":"Blackshear, S., Chursin, A., Danezis, G., Kichidis, A., Kokoris Kogias, E., Li, X., … Zhang, L. (2024). Sui Lutris: A blockchain combining broadcast and consensus. In <i>Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security</i> (pp. 2606–2620). Salt Lake City, UT, United States: ACM. <a href=\"https://doi.org/10.1145/3658644.3670286\">https://doi.org/10.1145/3658644.3670286</a>","ista":"Blackshear S, Chursin A, Danezis G, Kichidis A, Kokoris Kogias E, Li X, Logan M, Menon A, Nowacki T, Sonnino A, Williams B, Zhang L. 2024. Sui Lutris: A blockchain combining broadcast and consensus. Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security. CCS: Conference on Computer and Communications Security, 2606–2620.","ieee":"S. Blackshear <i>et al.</i>, “Sui Lutris: A blockchain combining broadcast and consensus,” in <i>Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security</i>, Salt Lake City, UT, United States, 2024, pp. 2606–2620.","chicago":"Blackshear, Sam, Andrey Chursin, George Danezis, Anastasios Kichidis, Eleftherios Kokoris Kogias, Xun Li, Mark Logan, et al. “Sui Lutris: A Blockchain Combining Broadcast and Consensus.” In <i>Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security</i>, 2606–20. ACM, 2024. <a href=\"https://doi.org/10.1145/3658644.3670286\">https://doi.org/10.1145/3658644.3670286</a>.","mla":"Blackshear, Sam, et al. “Sui Lutris: A Blockchain Combining Broadcast and Consensus.” <i>Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security</i>, ACM, 2024, pp. 2606–20, doi:<a href=\"https://doi.org/10.1145/3658644.3670286\">10.1145/3658644.3670286</a>."},"article_processing_charge":"No","day":"09","publication":"Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security","user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","type":"conference","date_published":"2024-12-09T00:00:00Z","external_id":{"isi":["001436367300178"],"arxiv":["2310.18042"]},"_id":"18957","date_created":"2025-01-29T12:42:21Z","year":"2024","doi":"10.1145/3658644.3670286","date_updated":"2025-09-09T12:13:55Z","status":"public","arxiv":1,"oa":1,"publisher":"ACM","abstract":[{"lang":"eng","text":"Sui Lutris is the first smart-contract platform to sustainably achieve sub-second finality. It achieves this significant decrease by employing consensusless agreement not only for simple payments but for a large variety of transactions. Unlike prior work, Sui Lutris neither compromises expressiveness nor throughput and can run perpetually without restarts. Sui Lutris achieves this by safely integrating consensuless agreement with a high-throughput consensus protocol that is invoked out of the critical finality path but ensures that when a transaction is at risk of inconsistent concurrent accesses, its settlement is delayed until the total ordering is resolved. Building such a hybrid architecture is especially delicate during reconfiguration events, where the system needs to preserve the safety of the consensusless path without compromising the long-term liveness of potentially misconfigured clients. We thus develop a novel reconfiguration protocol, the first to provably show the safe and efficient reconfiguration of a consensusless blockchain. Sui Lutris is currently running in production and underpins the Sui smart-contract platform. Combined with the use of Objects instead of accounts it enables the safe execution of smart contracts that expose objects as a first-class resource. In our experiments Sui Lutris achieves latency lower than 0.5 seconds for throughput up to 5,000 certificates per second (150k ops/s with transaction blocks), compared to the state-of-the-art real-world consensus latencies of 3 seconds. Furthermore, it gracefully handles validators crash-recovery and does not suffer visible performance degradation during reconfiguration."}],"title":"Sui Lutris: A blockchain combining broadcast and consensus","quality_controlled":"1","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2310.18042","open_access":"1"}],"publication_identifier":{"isbn":["9798400706363"]}},{"year":"2024","language":[{"iso":"eng"}],"page":"2247-2302","oa_version":"None","date_created":"2025-01-29T13:09:28Z","date_published":"2024-04-18T00:00:00Z","_id":"18958","publication_status":"published","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"journal_article","department":[{"_id":"RoSe"}],"author":[{"full_name":"Hainzl, Christian","last_name":"Hainzl","first_name":"Christian"},{"first_name":"Benjamin","full_name":"Schlein, Benjamin","last_name":"Schlein"},{"full_name":"Seiringer, Robert","last_name":"Seiringer","orcid":"0000-0002-6781-0521","id":"4AFD0470-F248-11E8-B48F-1D18A9856A87","first_name":"Robert"},{"first_name":"Simone","last_name":"Warzel","full_name":"Warzel, Simone"}],"month":"04","volume":20,"publication":"Oberwolfach Reports","intvolume":"        20","article_processing_charge":"No","day":"18","acknowledgement":"The MFO and the workshop organizers would like to thank the National Science Foundation for supporting the participation of junior researchers in the workshop by the grant DMS-2230648, “US Junior Oberwolfach Fellows”.","publication_identifier":{"eissn":["1660-8941"],"issn":["1660-8933"]},"citation":{"mla":"Hainzl, Christian, et al. “Many-Body Quantum Systems.” <i>Oberwolfach Reports</i>, vol. 20, no. 3, EMS Press, 2024, pp. 2247–302, doi:<a href=\"https://doi.org/10.4171/owr/2023/39\">10.4171/owr/2023/39</a>.","chicago":"Hainzl, Christian, Benjamin Schlein, Robert Seiringer, and Simone Warzel. “Many-Body Quantum Systems.” <i>Oberwolfach Reports</i>. EMS Press, 2024. <a href=\"https://doi.org/10.4171/owr/2023/39\">https://doi.org/10.4171/owr/2023/39</a>.","ieee":"C. Hainzl, B. Schlein, R. Seiringer, and S. Warzel, “Many-body quantum systems,” <i>Oberwolfach Reports</i>, vol. 20, no. 3. EMS Press, pp. 2247–2302, 2024.","ista":"Hainzl C, Schlein B, Seiringer R, Warzel S. 2024. Many-body quantum systems. Oberwolfach Reports. 20(3), 2247–2302.","apa":"Hainzl, C., Schlein, B., Seiringer, R., &#38; Warzel, S. (2024). Many-body quantum systems. <i>Oberwolfach Reports</i>. EMS Press. <a href=\"https://doi.org/10.4171/owr/2023/39\">https://doi.org/10.4171/owr/2023/39</a>","ama":"Hainzl C, Schlein B, Seiringer R, Warzel S. Many-body quantum systems. <i>Oberwolfach Reports</i>. 2024;20(3):2247-2302. doi:<a href=\"https://doi.org/10.4171/owr/2023/39\">10.4171/owr/2023/39</a>","short":"C. Hainzl, B. Schlein, R. Seiringer, S. Warzel, Oberwolfach Reports 20 (2024) 2247–2302."},"issue":"3","publisher":"EMS Press","quality_controlled":"1","title":"Many-body quantum systems","abstract":[{"text":"This workshop brought together experts on the analysis of quantum many-body problems and quantum statistical mechanics, with the goal of discussing the state-of-the-art of the field, recent developments as well as challenges for the future. The main topics of discussion concerned the equilibrium and dynamical behavior of (bosonic or fermionic) quantum gases, quantum spin systems, as well as quantum field theory models like the Nelson or Fröhlich model.","lang":"eng"}],"article_type":"original","status":"public","doi":"10.4171/owr/2023/39","date_updated":"2025-01-29T13:19:06Z"},{"year":"2024","file_date_updated":"2025-01-29T13:44:47Z","date_created":"2025-01-29T13:39:34Z","license":"https://creativecommons.org/licenses/by/4.0/","_id":"18961","date_published":"2024-07-01T00:00:00Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"journal_article","publication":"Proceedings on Privacy Enhancing Technologies","day":"01","article_processing_charge":"No","publication_identifier":{"issn":["2299-0984"]},"abstract":[{"lang":"eng","text":"Automated contact tracing (ACT) emerged as a promising measure to curb the spread of Covid-19. Users enable ACT on their smartphones to automatically record contacts with other users. If a user tests positive for the disease, they report their diagnosis to alert their contacts.\r\nDesigning effective ACT protocols is challenging since they need to be efficient and secure while also ensuring users' privacy. As ACT protocols necessarily leak some information by design, defining privacy is difficult. For example, a user cannot deny having met another user. Ideally, however, the user can plausibly deny everything else, in particular, when they met. We call this privacy property contact-time deniability.\r\nWhile some early works discussed contact-time deniability informally, it has received little attention since then. We investigate deniability from a rigorous, theoretical point of view and arrive at the following impossibility result:\r\nA decentralized protocol with unidirectional communication cannot be contact-time deniable and replay-secure. This holds even if malicious users treat smartphones as black-boxes.\r\n Unidirectional protocols are usually very efficient and many proposals are unidirectional, e.g., the widely-deployed Google-Apple Exposure Notifications. So the impossibility result considerably constrains the design space of efficient, secure, and private ACT protocols. However, it can also be used as a guide; we discuss several possibilities to achieve contact-time deniability in practice."}],"quality_controlled":"1","title":"Deniability in automated contact tracing: Impossibilities and possibilities","publisher":"Privacy Enhancing Technologies Symposium Advisory Board","file":[{"file_name":"2024_ProcPrivacyEnhTech_Guenther.pdf","date_created":"2025-01-29T13:44:47Z","relation":"main_file","access_level":"open_access","creator":"dernst","file_size":611567,"success":1,"checksum":"348ed6adcf6ad2f925227bde1758cae6","file_id":"18962","content_type":"application/pdf","date_updated":"2025-01-29T13:44:47Z"}],"article_type":"original","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"oa":1,"doi":"10.56553/popets-2024-0134","date_updated":"2025-04-15T08:16:04Z","status":"public","project":[{"grant_number":"F8509","name":"Security and Privacy by Design for Complex Systems","_id":"34a34d57-11ca-11ed-8bc3-a2688a8724e1"}],"language":[{"iso":"eng"}],"oa_version":"Published Version","page":"636-648","department":[{"_id":"KrPi"},{"_id":"GradSch"}],"author":[{"id":"ec98511c-eb8e-11eb-b029-edd25d7271a1","first_name":"Christoph Ullrich","full_name":"Günther, Christoph Ullrich","last_name":"Günther"},{"orcid":"0000-0002-9139-1654","full_name":"Pietrzak, Krzysztof Z","last_name":"Pietrzak","first_name":"Krzysztof Z","id":"3E04A7AA-F248-11E8-B48F-1D18A9856A87"}],"conference":{"end_date":"2024-07-20","start_date":"2024-07-15","location":"Bristol, UK/Virtual","name":"PETs: Privacy Enhancing Technologies Symposium "},"publication_status":"published","month":"07","volume":2024,"ddc":["000"],"intvolume":"      2024","acknowledgement":"We thank Raluca-Georgia Diugan for her initial contributions and support afterward.\r\nThis research was funded in whole or in part by the Austrian Science Fund (FWF) 10.55776/F85.","citation":{"ieee":"C. U. Günther and K. Z. Pietrzak, “Deniability in automated contact tracing: Impossibilities and possibilities,” <i>Proceedings on Privacy Enhancing Technologies</i>, vol. 2024, no. 4. Privacy Enhancing Technologies Symposium Advisory Board, pp. 636–648, 2024.","chicago":"Günther, Christoph Ullrich, and Krzysztof Z Pietrzak. “Deniability in Automated Contact Tracing: Impossibilities and Possibilities.” <i>Proceedings on Privacy Enhancing Technologies</i>. Privacy Enhancing Technologies Symposium Advisory Board, 2024. <a href=\"https://doi.org/10.56553/popets-2024-0134\">https://doi.org/10.56553/popets-2024-0134</a>.","mla":"Günther, Christoph Ullrich, and Krzysztof Z. Pietrzak. “Deniability in Automated Contact Tracing: Impossibilities and Possibilities.” <i>Proceedings on Privacy Enhancing Technologies</i>, vol. 2024, no. 4, Privacy Enhancing Technologies Symposium Advisory Board, 2024, pp. 636–48, doi:<a href=\"https://doi.org/10.56553/popets-2024-0134\">10.56553/popets-2024-0134</a>.","ista":"Günther CU, Pietrzak KZ. 2024. Deniability in automated contact tracing: Impossibilities and possibilities. Proceedings on Privacy Enhancing Technologies. 2024(4), 636–648.","ama":"Günther CU, Pietrzak KZ. Deniability in automated contact tracing: Impossibilities and possibilities. <i>Proceedings on Privacy Enhancing Technologies</i>. 2024;2024(4):636-648. doi:<a href=\"https://doi.org/10.56553/popets-2024-0134\">10.56553/popets-2024-0134</a>","apa":"Günther, C. U., &#38; Pietrzak, K. Z. (2024). Deniability in automated contact tracing: Impossibilities and possibilities. <i>Proceedings on Privacy Enhancing Technologies</i>. Bristol, UK/Virtual: Privacy Enhancing Technologies Symposium Advisory Board. <a href=\"https://doi.org/10.56553/popets-2024-0134\">https://doi.org/10.56553/popets-2024-0134</a>","short":"C.U. Günther, K.Z. Pietrzak, Proceedings on Privacy Enhancing Technologies 2024 (2024) 636–648."},"corr_author":"1","issue":"4","OA_type":"gold","has_accepted_license":"1","OA_place":"publisher"},{"acknowledgement":"Yanwei Fu is the corresponding authour. Yanwei Fu is with School of Data Science, Fudan University, Shanghai Key Lab of Intelligent Information Processing, Fudan University, and Fudan ISTBI-ZJNU Algorithm Centre for Brain-inspired Intelligence, Zhejiang Normal University, Jinhua, China.","author":[{"first_name":"Ke","last_name":"Fan","full_name":"Fan, Ke"},{"full_name":"Bai, Zechen","last_name":"Bai","first_name":"Zechen"},{"first_name":"Tianjun","full_name":"Xiao, Tianjun","last_name":"Xiao"},{"full_name":"He, Tong","last_name":"He","first_name":"Tong"},{"first_name":"Max","full_name":"Horn, Max","last_name":"Horn"},{"last_name":"Fu","full_name":"Fu, Yanwei","first_name":"Yanwei"},{"full_name":"Locatello, Francesco","last_name":"Locatello","orcid":"0000-0002-4850-0683","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"},{"first_name":"Zheng","last_name":"Zhang","full_name":"Zhang, Zheng"}],"department":[{"_id":"FrLo"}],"conference":{"location":"Seattle, WA, United States","start_date":"2024-06-16","end_date":"2024-06-22","name":"CVPR: Conference on Computer Vision and Pattern Recognition"},"isi":1,"publication_status":"published","month":"06","language":[{"iso":"eng"}],"oa_version":"Preprint","OA_place":"repository","related_material":{"link":[{"relation":"software","url":"https://kfan21.github.io/AdaSlot/"}]},"citation":{"ista":"Fan K, Bai Z, Xiao T, He T, Horn M, Fu Y, Locatello F, Zhang Z. 2024. Adaptive slot attention: Object discovery with dynamic slot number. 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. CVPR: Conference on Computer Vision and Pattern Recognition.","mla":"Fan, Ke, et al. “Adaptive Slot Attention: Object Discovery with Dynamic Slot Number.” <i>2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, IEEE, 2024, doi:<a href=\"https://doi.org/10.1109/cvpr52733.2024.02176\">10.1109/cvpr52733.2024.02176</a>.","ieee":"K. Fan <i>et al.</i>, “Adaptive slot attention: Object discovery with dynamic slot number,” in <i>2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>, Seattle, WA, United States, 2024.","chicago":"Fan, Ke, Zechen Bai, Tianjun Xiao, Tong He, Max Horn, Yanwei Fu, Francesco Locatello, and Zheng Zhang. “Adaptive Slot Attention: Object Discovery with Dynamic Slot Number.” In <i>2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>. IEEE, 2024. <a href=\"https://doi.org/10.1109/cvpr52733.2024.02176\">https://doi.org/10.1109/cvpr52733.2024.02176</a>.","short":"K. Fan, Z. Bai, T. Xiao, T. He, M. Horn, Y. Fu, F. Locatello, Z. Zhang, in:, 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, 2024.","apa":"Fan, K., Bai, Z., Xiao, T., He, T., Horn, M., Fu, Y., … Zhang, Z. (2024). Adaptive slot attention: Object discovery with dynamic slot number. In <i>2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>. Seattle, WA, United States: IEEE. <a href=\"https://doi.org/10.1109/cvpr52733.2024.02176\">https://doi.org/10.1109/cvpr52733.2024.02176</a>","ama":"Fan K, Bai Z, Xiao T, et al. Adaptive slot attention: Object discovery with dynamic slot number. In: <i>2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition</i>. IEEE; 2024. doi:<a href=\"https://doi.org/10.1109/cvpr52733.2024.02176\">10.1109/cvpr52733.2024.02176</a>"},"OA_type":"green","day":"15","article_processing_charge":"No","user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","type":"conference","publication":"2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition","date_created":"2025-01-29T14:27:39Z","external_id":{"isi":["001342515506043"],"arxiv":["2406.09196"]},"_id":"18964","date_published":"2024-06-15T00:00:00Z","year":"2024","arxiv":1,"status":"public","date_updated":"2025-09-09T12:15:17Z","doi":"10.1109/cvpr52733.2024.02176","oa":1,"quality_controlled":"1","title":"Adaptive slot attention: Object discovery with dynamic slot number","abstract":[{"text":"Object-centric learning (OCL) extracts the representation of objects with slots, offering an exceptional blend of flexibility and interpretability for abstracting low-level perceptual features. A widely adopted method within OCL is slot attention, which utilizes attention mechanisms to iteratively refine slot representations. However, a major draw-back of most object-centric models, including slot attention, is their reliance on predefining the number of slots. This not only necessitates prior knowledge of the dataset but also overlooks the inherent variability in the number of objects present in each instance. To overcome this fundamental limitation, we present a novel complexity-aware object auto-encoder framework. Within this framework, we introduce an adaptive slot attention (AdaSlot) mecha-nism that dynamically determines the optimal number of slots based on the content of the data. This is achieved by proposing a discrete slot sampling module that is responsible for selecting an appropriate number of slots from a candidate list. Furthermore, we introduce a masked slot decoder that suppresses unselected slots during the decoding process. Our framework, tested extensively on object discovery tasks with various datasets, shows performance matching or exceeding top fixed-slot models. Moreover, our analysis substantiates that our method exhibits the capability to dynamically adapt the slot number according to each instance's complexity, offering the potential for further exploration in slot attention research. Project will be available at https://kfan21.github.io/AdaSlot/","lang":"eng"}],"publisher":"IEEE","publication_identifier":{"eisbn":["9798350353006"]},"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2406.09196","open_access":"1"}]},{"intvolume":"        20","ddc":["570"],"publication_status":"published","author":[{"last_name":"Petrova","full_name":"Petrova, Olga","first_name":"Olga","id":"5D8C9660-5D49-11EA-8188-567B3DDC885E"},{"first_name":"Sergey A","full_name":"Trushin, Sergey A","last_name":"Trushin"},{"first_name":"Thi Kim Oanh","full_name":"Nguyen, Thi Kim Oanh","last_name":"Nguyen"},{"first_name":"Mark","last_name":"Ostroot","full_name":"Ostroot, Mark"},{"full_name":"Schellenberg, Matthew","last_name":"Schellenberg","first_name":"Matthew"},{"first_name":"Graham","full_name":"Johnson, Graham","last_name":"Johnson"},{"first_name":"Eugenia","full_name":"Trushina, Eugenia","last_name":"Trushina"},{"first_name":"Leonid A","id":"338D39FE-F248-11E8-B48F-1D18A9856A87","full_name":"Sazanov, Leonid A","last_name":"Sazanov","orcid":"0000-0002-0977-7989"}],"department":[{"_id":"LeSa"}],"month":"12","volume":20,"language":[{"iso":"eng"}],"oa_version":"Published Version","OA_place":"publisher","has_accepted_license":"1","article_number":"e085971","citation":{"ista":"Petrova O, Trushin SA, Nguyen TKO, Ostroot M, Schellenberg M, Johnson G, Trushina E, Sazanov LA. 2024. Structure‐activity relationship study of neuroprotective complex I inhibitor CP2, Wiley,p.","chicago":"Petrova, Olga, Sergey A Trushin, Thi Kim Oanh Nguyen, Mark Ostroot, Matthew Schellenberg, Graham Johnson, Eugenia Trushina, and Leonid A Sazanov. <i>Structure‐activity Relationship Study of Neuroprotective Complex I Inhibitor CP2</i>. <i>Alzheimer’s &#38; Dementia</i>. Vol. 20. Wiley, 2024. <a href=\"https://doi.org/10.1002/alz.085971\">https://doi.org/10.1002/alz.085971</a>.","ieee":"O. Petrova <i>et al.</i>, <i>Structure‐activity relationship study of neuroprotective complex I inhibitor CP2</i>, vol. 20, no. S6. Wiley, 2024.","mla":"Petrova, Olga, et al. “Structure‐activity Relationship Study of Neuroprotective Complex I Inhibitor CP2.” <i>Alzheimer’s &#38; Dementia</i>, vol. 20, no. S6, e085971, Wiley, 2024, doi:<a href=\"https://doi.org/10.1002/alz.085971\">10.1002/alz.085971</a>.","short":"O. Petrova, S.A. Trushin, T.K.O. Nguyen, M. Ostroot, M. Schellenberg, G. Johnson, E. Trushina, L.A. Sazanov, Structure‐activity Relationship Study of Neuroprotective Complex I Inhibitor CP2, Wiley, 2024.","ama":"Petrova O, Trushin SA, Nguyen TKO, et al. <i>Structure‐activity Relationship Study of Neuroprotective Complex I Inhibitor CP2</i>. Vol 20. Wiley; 2024. doi:<a href=\"https://doi.org/10.1002/alz.085971\">10.1002/alz.085971</a>","apa":"Petrova, O., Trushin, S. A., Nguyen, T. K. O., Ostroot, M., Schellenberg, M., Johnson, G., … Sazanov, L. A. (2024). <i>Structure‐activity relationship study of neuroprotective complex I inhibitor CP2</i>. <i>Alzheimer’s &#38; Dementia</i> (Vol. 20). Wiley. <a href=\"https://doi.org/10.1002/alz.085971\">https://doi.org/10.1002/alz.085971</a>"},"issue":"S6","OA_type":"hybrid","article_processing_charge":"Yes (in subscription journal)","day":"01","type":"other_academic_publication","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication":"Alzheimer's & Dementia","date_created":"2025-01-29T15:21:40Z","date_published":"2024-12-01T00:00:00Z","_id":"18967","year":"2024","file_date_updated":"2025-01-29T15:24:50Z","doi":"10.1002/alz.085971","status":"public","date_updated":"2025-01-29T15:29:22Z","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"oa":1,"publisher":"Wiley","title":"Structure‐activity relationship study of neuroprotective complex I inhibitor CP2","abstract":[{"text":"Background: We identified small molecule tricyclic pyrone compound CP2 as a mild mitochondrial complex I (MCI) inhibitor that induces neuroprotection in multiple mouse models of AD. One of the major concerns while targeting mitochondria is the production of reactive oxygen species (ROS). CP2 consists of two diastereoisomers, D1 and D2, with distinct activity and toxicity profiles. This study was designed to understand how structure of D1 and D2 affects their binding to MCI and the consequential impact on ROS production.\r\n\r\nMethod: The X-ray crystallography and cryo-electron microscopy (cryo-EM) at global resolution of 3.25-3.27Å were employed to identify the molecular structure of D1 and D2 and the D1 binding to the isolated ovine MCI. The assessment of the MCI inhibition and the extent of ROS generation were done in isolated MCI and human neuroblastoma MC65 cells using flow cytometry, a Seahorse extracellular flux analyzer, and the kinetic studies.\r\n\r\nResult: In the closed conformation of MCI, D1 selectively binds to the deep Quinone-site (Qd) but not to the shallow Q-site (Qs), sharing the same binding pocket as rotenone. In the open MCI state, D1 exclusively binds to the Qs in contrast to rotenone, which binds Qd and Qs in both closed and open states. At the same concentrations, D1 inhibits respiration to a greater extent compared to D2 (5:1 ratio) and produces higher level of ROS.\r\n\r\nConclusion:Cryo-EM unambiguously identified binding of D1 to both the Qd and Qs sites, contingent upon the conformational state of MCI. In contrast to rotenone, D1 binds Qd only in the closed conformation during catalytic cycle, leading to mild inhibition. Superimposing X-ray crystallography data of D1 and D2 onto cryo-EM data suggests that the orientation of the methyl group in D2 induces a flatter conformation, resulting in lower binding affinity to MCI, which correlates with lower inhibition and toxicity compared to D1. At physiologically relevant concentrations, CP2 (D1:D2 = 1:1) demonstrates low MCI inhibition yielding negligible ROS levels. This observation provides new insight into the absence of toxicity associated with CP2 treatment in vivo, further highlighting feasibility for the development of safe and efficacious MCI inhibitors.","lang":"eng"}],"quality_controlled":"1","file":[{"file_name":"2024_AlzheimerDementia_Petrova.pdf","access_level":"open_access","file_size":70870,"creator":"dernst","relation":"main_file","date_created":"2025-01-29T15:24:50Z","content_type":"application/pdf","checksum":"e914bd5f3a701659ab79d497a122811f","file_id":"18968","success":1,"date_updated":"2025-01-29T15:24:50Z"}],"publication_identifier":{"eissn":["1552-5279"],"issn":["1552-5260"]}},{"license":"https://creativecommons.org/licenses/by-sa/4.0/","date_created":"2025-01-29T15:34:22Z","_id":"18970","date_published":"2024-05-05T00:00:00Z","year":"2024","day":"05","article_processing_charge":"No","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"journal_article","publication":"Oberwolfach Reports","abstract":[{"text":"Given a smooth projective curve C, nonabelian Hodge theory gives a diffeomorphism between two different moduli spaces associated to C. The first is the moduli space of Higgs bundles on C of rank n, which is equipped with the structure of an algebraic completely integrable Hamiltonian system. The second is the character variety of representations of the fundamental group of C into GL(n). In 2012, de Cataldo, Hausel, and Migliorini [1] proposed the P=W conjecture which identifies the perverse filtration on the cohomology of the Higgs moduli space with the weight filtration on the cohomology of the character variety. Recently, in 2022, two independent proofs of the P=W Conjecture appeared, in work of Maulik &Shen [2] and Hausel, Mellit, Minets &Schiffmann [6]. The aim of the Arbeitsgemeinschaft was to understand the P=W Conjecture and these two recent proofs.","lang":"eng"}],"title":"Arbeitsgemeinschaft: Geometry and representation theory around the P=W conjecture","quality_controlled":"1","publisher":"EMS Press","publication_identifier":{"eissn":["1660-8941"],"issn":["1660-8933"]},"main_file_link":[{"url":"https://doi.org/10.4171/owr/2024/16","open_access":"1"}],"status":"public","doi":"10.4171/owr/2024/16","date_updated":"2025-01-29T15:39:55Z","article_type":"original","tmp":{"image":"/images/cc_by_sa.png","short":"CC BY-SA (4.0)","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)"},"oa":1,"language":[{"iso":"eng"}],"oa_version":"Published Version","page":"949-1004","ddc":["500"],"intvolume":"        21","acknowledgement":"The MFO and the workshop organizers would like to thank the\r\nNational Science Foundation for supporting the participation of junior researchers\r\nby the grant DMS-2230648, “US Junior Oberwolfach Fellows”. Moreover, the\r\nMFO and the workshop organizers would like to thank the Oberwolfach Foundation for supporting the participation of junior researchers in the Arbeitsgemeinschaft.","department":[{"_id":"TaHa"}],"author":[{"id":"4A0666D8-F248-11E8-B48F-1D18A9856A87","first_name":"Tamás","orcid":"0000-0002-9582-2634","full_name":"Hausel, Tamás","last_name":"Hausel"},{"first_name":"Davesh","full_name":"Maulik, Davesh","last_name":"Maulik"},{"first_name":"Anton","last_name":"Mellit","full_name":"Mellit, Anton"},{"first_name":"Olivier","full_name":"Schiffmann, Olivier","last_name":"Schiffmann"},{"full_name":"Shen, Junliang","last_name":"Shen","first_name":"Junliang"}],"publication_status":"published","volume":21,"month":"05","has_accepted_license":"1","citation":{"ista":"Hausel T, Maulik D, Mellit A, Schiffmann O, Shen J. 2024. Arbeitsgemeinschaft: Geometry and representation theory around the P=W conjecture. Oberwolfach Reports. 21(2), 949–1004.","mla":"Hausel, Tamás, et al. “Arbeitsgemeinschaft: Geometry and Representation Theory around the P=W Conjecture.” <i>Oberwolfach Reports</i>, vol. 21, no. 2, EMS Press, 2024, pp. 949–1004, doi:<a href=\"https://doi.org/10.4171/owr/2024/16\">10.4171/owr/2024/16</a>.","chicago":"Hausel, Tamás, Davesh Maulik, Anton Mellit, Olivier Schiffmann, and Junliang Shen. “Arbeitsgemeinschaft: Geometry and Representation Theory around the P=W Conjecture.” <i>Oberwolfach Reports</i>. EMS Press, 2024. <a href=\"https://doi.org/10.4171/owr/2024/16\">https://doi.org/10.4171/owr/2024/16</a>.","ieee":"T. Hausel, D. Maulik, A. Mellit, O. Schiffmann, and J. Shen, “Arbeitsgemeinschaft: Geometry and representation theory around the P=W conjecture,” <i>Oberwolfach Reports</i>, vol. 21, no. 2. EMS Press, pp. 949–1004, 2024.","short":"T. Hausel, D. Maulik, A. Mellit, O. Schiffmann, J. Shen, Oberwolfach Reports 21 (2024) 949–1004.","apa":"Hausel, T., Maulik, D., Mellit, A., Schiffmann, O., &#38; Shen, J. (2024). Arbeitsgemeinschaft: Geometry and representation theory around the P=W conjecture. <i>Oberwolfach Reports</i>. EMS Press. <a href=\"https://doi.org/10.4171/owr/2024/16\">https://doi.org/10.4171/owr/2024/16</a>","ama":"Hausel T, Maulik D, Mellit A, Schiffmann O, Shen J. Arbeitsgemeinschaft: Geometry and representation theory around the P=W conjecture. <i>Oberwolfach Reports</i>. 2024;21(2):949-1004. doi:<a href=\"https://doi.org/10.4171/owr/2024/16\">10.4171/owr/2024/16</a>"},"OA_type":"hybrid","issue":"2","OA_place":"publisher"},{"day":"30","article_processing_charge":"No","publication":"Proceedings of the 41st International Conference on Machine Learning","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","_id":"18971","external_id":{"arxiv":["2402.13368"]},"date_published":"2024-07-30T00:00:00Z","date_created":"2025-01-30T07:21:57Z","year":"2024","status":"public","date_updated":"2025-01-30T07:23:10Z","arxiv":1,"oa":1,"quality_controlled":"1","abstract":[{"lang":"eng","text":"Models prone to spurious correlations in training data often produce brittle predictions and introduce unintended biases. Addressing this challenge typically involves methods relying on prior knowledge and group annotation to remove spurious correlations, which may not be readily available in many applications. In this paper, we establish a novel connection between unsupervised object-centric learning and mitigation of spurious correlations. Instead of directly inferring subgroups with varying correlations with labels, our approach focuses on discovering concepts: discrete ideas that are shared across input samples. Leveraging existing object-centric representation learning, we introduce CoBalT: a concept balancing technique that effectively mitigates spurious correlations without requiring human labeling of subgroups. Evaluation across the benchmark datasets for sub-population shifts demonstrate superior or competitive performance compared state-of-the-art baselines, without the need for group annotation. Code is available at https://github.com/rarefin/CoBalT"}],"title":"Unsupervised concept discovery mitigates spurious correlations","publisher":"ML Research Press","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2402.13368","open_access":"1"}],"publication_identifier":{"eissn":["2640-3498"]},"acknowledgement":"We acknowledge the support of the Canada CIFAR AI Chair Program and IVADO. We thank Mila and Compute Canada for providing computational resources.\r\n","scopus_import":"1","intvolume":"       235","volume":235,"month":"07","conference":{"name":"ICML: International Conference on Machine Learning","start_date":"2024-07-21","location":"Vienna, Austria","end_date":"2024-07-27"},"author":[{"full_name":"Arefin, Rifat","last_name":"Arefin","first_name":"Rifat"},{"first_name":"Yan","last_name":"Zhang","full_name":"Zhang, Yan"},{"full_name":"Baratin, Aristide","last_name":"Baratin","first_name":"Aristide"},{"first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683"},{"first_name":"Irina","last_name":"Rish","full_name":"Rish, Irina"},{"last_name":"Liu","full_name":"Liu, Dianbo","first_name":"Dianbo"},{"last_name":"Kawaguchi","full_name":"Kawaguchi, Kenji","first_name":"Kenji"}],"department":[{"_id":"FrLo"}],"publication_status":"published","oa_version":"Preprint","page":"1672-1688","language":[{"iso":"eng"}],"OA_place":"repository","related_material":{"link":[{"relation":"software","url":"https://github.com/rarefin/CoBalT"}]},"alternative_title":["PMLR"],"OA_type":"green","citation":{"ama":"Arefin R, Zhang Y, Baratin A, et al. Unsupervised concept discovery mitigates spurious correlations. In: <i>Proceedings of the 41st International Conference on Machine Learning</i>. Vol 235. ML Research Press; 2024:1672-1688.","apa":"Arefin, R., Zhang, Y., Baratin, A., Locatello, F., Rish, I., Liu, D., &#38; Kawaguchi, K. (2024). Unsupervised concept discovery mitigates spurious correlations. In <i>Proceedings of the 41st International Conference on Machine Learning</i> (Vol. 235, pp. 1672–1688). Vienna, Austria: ML Research Press.","short":"R. Arefin, Y. Zhang, A. Baratin, F. Locatello, I. Rish, D. Liu, K. Kawaguchi, in:, Proceedings of the 41st International Conference on Machine Learning, ML Research Press, 2024, pp. 1672–1688.","chicago":"Arefin, Rifat, Yan Zhang, Aristide Baratin, Francesco Locatello, Irina Rish, Dianbo Liu, and Kenji Kawaguchi. “Unsupervised Concept Discovery Mitigates Spurious Correlations.” In <i>Proceedings of the 41st International Conference on Machine Learning</i>, 235:1672–88. ML Research Press, 2024.","ieee":"R. Arefin <i>et al.</i>, “Unsupervised concept discovery mitigates spurious correlations,” in <i>Proceedings of the 41st International Conference on Machine Learning</i>, Vienna, Austria, 2024, vol. 235, pp. 1672–1688.","mla":"Arefin, Rifat, et al. “Unsupervised Concept Discovery Mitigates Spurious Correlations.” <i>Proceedings of the 41st International Conference on Machine Learning</i>, vol. 235, ML Research Press, 2024, pp. 1672–88.","ista":"Arefin R, Zhang Y, Baratin A, Locatello F, Rish I, Liu D, Kawaguchi K. 2024. Unsupervised concept discovery mitigates spurious correlations. Proceedings of the 41st International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 235, 1672–1688."}},{"acknowledgement":"The authors were partially supported by the 2019 LopezLoreta prize, and they would like to thank (in alphabetical order) Grigorios Chrysos, Simone Maria Giancola, Mahyar\r\nJafari Nodeh, Christoph Lampert, Marco Miani, GuanWen Qiu, and Peter Sukenık for helpful discussions.","scopus_import":"1","intvolume":"       235","month":"07","volume":235,"department":[{"_id":"MaMo"}],"author":[{"first_name":"Simone","id":"ca726dda-de17-11ea-bc14-f9da834f63aa","last_name":"Bombari","full_name":"Bombari, Simone"},{"id":"27EB676C-8706-11E9-9510-7717E6697425","first_name":"Marco","full_name":"Mondelli, Marco","last_name":"Mondelli","orcid":"0000-0002-3242-7020"}],"conference":{"start_date":"2024-07-21","location":"Vienna, Austria","end_date":"2024-07-27","name":"ICML: International Conference on Machine Learning"},"publication_status":"published","oa_version":"Preprint","page":"4267-4299","language":[{"iso":"eng"}],"OA_place":"repository","alternative_title":["PMLR"],"OA_type":"green","citation":{"apa":"Bombari, S., &#38; Mondelli, M. (2024). How spurious features are memorized: Precise analysis for random and NTK features. In <i>41st International Conference on Machine Learning</i> (Vol. 235, pp. 4267–4299). Vienna, Austria: ML Research Press.","ama":"Bombari S, Mondelli M. How spurious features are memorized: Precise analysis for random and NTK features. In: <i>41st International Conference on Machine Learning</i>. Vol 235. ML Research Press; 2024:4267-4299.","short":"S. Bombari, M. Mondelli, in:, 41st International Conference on Machine Learning, ML Research Press, 2024, pp. 4267–4299.","mla":"Bombari, Simone, and Marco Mondelli. “How Spurious Features Are Memorized: Precise Analysis for Random and NTK Features.” <i>41st International Conference on Machine Learning</i>, vol. 235, ML Research Press, 2024, pp. 4267–99.","ieee":"S. Bombari and M. Mondelli, “How spurious features are memorized: Precise analysis for random and NTK features,” in <i>41st International Conference on Machine Learning</i>, Vienna, Austria, 2024, vol. 235, pp. 4267–4299.","chicago":"Bombari, Simone, and Marco Mondelli. “How Spurious Features Are Memorized: Precise Analysis for Random and NTK Features.” In <i>41st International Conference on Machine Learning</i>, 235:4267–99. ML Research Press, 2024.","ista":"Bombari S, Mondelli M. 2024. How spurious features are memorized: Precise analysis for random and NTK features. 41st International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 235, 4267–4299."},"corr_author":"1","day":"30","article_processing_charge":"No","publication":"41st International Conference on Machine Learning","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","_id":"18972","external_id":{"arxiv":["2305.12100"]},"date_published":"2024-07-30T00:00:00Z","date_created":"2025-01-30T07:29:47Z","year":"2024","date_updated":"2025-04-15T07:50:12Z","status":"public","project":[{"name":"Prix Lopez-Loretta 2019 - Marco Mondelli","_id":"059876FA-7A3F-11EA-A408-12923DDC885E"}],"arxiv":1,"oa":1,"abstract":[{"text":"Deep learning models are known to overfit and memorize spurious features in the training dataset. While numerous empirical studies have aimed at understanding this phenomenon, a rigorous theoretical framework to quantify it is still missing. In this paper, we consider spurious features that are uncorrelated with the learning task, and we provide a precise characterization of how they are memorized via two separate terms: (i) the stability of the model with respect to individual training samples, and (ii) the feature alignment between the spurious pattern and the full sample. While the first term is well established in learning theory and it is connected to the generalization error in classical work, the second one is, to the best of our knowledge, novel. Our key technical result gives a precise characterization of the feature alignment for the two prototypical settings of random features (RF) and neural tangent kernel (NTK) regression. We prove that the memorization of spurious features weakens as the generalization capability increases and, through the analysis of the feature alignment, we unveil the role of the model and of its activation function. Numerical experiments show the predictive power of our theory on standard datasets (MNIST, CIFAR-10).","lang":"eng"}],"quality_controlled":"1","title":"How spurious features are memorized: Precise analysis for random and NTK features","publisher":"ML Research Press","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2305.12100"}],"publication_identifier":{"eissn":["2640-3498"]}},{"day":"30","article_processing_charge":"No","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","publication":"41st International Conference on Machine Learning","date_created":"2025-01-30T07:35:49Z","_id":"18973","external_id":{"arxiv":["2402.02969"]},"date_published":"2024-07-30T00:00:00Z","year":"2024","arxiv":1,"date_updated":"2025-04-15T07:50:12Z","status":"public","project":[{"name":"Prix Lopez-Loretta 2019 - Marco Mondelli","_id":"059876FA-7A3F-11EA-A408-12923DDC885E"}],"oa":1,"abstract":[{"lang":"eng","text":"Understanding the reasons behind the exceptional success of transformers requires a better analysis of why attention layers are suitable for NLP tasks. In particular, such tasks require predictive models to capture contextual meaning which often depends on one or few words, even if the sentence is long. Our work studies this key property, dubbed word sensitivity (WS), in the prototypical setting of random features. We show that attention layers enjoy high WS, namely, there exists a vector in the space of embeddings that largely perturbs the random attention features map. The argument critically exploits the role of the softmax in the attention layer, highlighting its benefit compared to other activations (e.g., ReLU). In contrast, the WS of standard random features is of order 1/n−−√, n being the number of words in the textual sample, and thus it decays with the length of the context. We then translate these results on the word sensitivity into generalization bounds: due to their low WS, random features provably cannot learn to distinguish between two sentences that differ only in a single word; in contrast, due to their high WS, random attention features have higher generalization capabilities. We validate our theoretical results with experimental evidence over the BERT-Base word embeddings of the imdb review dataset."}],"title":"Towards understanding the word sensitivity of attention layers: A study via random features","quality_controlled":"1","publisher":"ML Research Press","publication_identifier":{"eissn":["2640-3498"]},"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2402.02969","open_access":"1"}],"scopus_import":"1","intvolume":"       235","acknowledgement":"The authors were partially supported by the 2019 LopezLoreta prize, and they would like to thank Mohammad Hossein Amani, Lorenzo Beretta, and Clement Rebuffel for helpful discussions.","author":[{"id":"ca726dda-de17-11ea-bc14-f9da834f63aa","first_name":"Simone","full_name":"Bombari, Simone","last_name":"Bombari"},{"orcid":"0000-0002-3242-7020","full_name":"Mondelli, Marco","last_name":"Mondelli","first_name":"Marco","id":"27EB676C-8706-11E9-9510-7717E6697425"}],"conference":{"name":"ICML: International Conference on Machine Learning","start_date":"2024-07-21","location":"Vienna, Austria","end_date":"2024-07-27"},"department":[{"_id":"MaMo"}],"publication_status":"published","volume":235,"month":"07","language":[{"iso":"eng"}],"oa_version":"Preprint","page":"4300-4328","OA_place":"repository","alternative_title":["PMLR"],"citation":{"ama":"Bombari S, Mondelli M. Towards understanding the word sensitivity of attention layers: A study via random features. In: <i>41st International Conference on Machine Learning</i>. Vol 235. ML Research Press; 2024:4300-4328.","apa":"Bombari, S., &#38; Mondelli, M. (2024). Towards understanding the word sensitivity of attention layers: A study via random features. In <i>41st International Conference on Machine Learning</i> (Vol. 235, pp. 4300–4328). Vienna, Austria: ML Research Press.","short":"S. Bombari, M. Mondelli, in:, 41st International Conference on Machine Learning, ML Research Press, 2024, pp. 4300–4328.","chicago":"Bombari, Simone, and Marco Mondelli. “Towards Understanding the Word Sensitivity of Attention Layers: A Study via Random Features.” In <i>41st International Conference on Machine Learning</i>, 235:4300–4328. ML Research Press, 2024.","ieee":"S. Bombari and M. Mondelli, “Towards understanding the word sensitivity of attention layers: A study via random features,” in <i>41st International Conference on Machine Learning</i>, Vienna, Austria, 2024, vol. 235, pp. 4300–4328.","mla":"Bombari, Simone, and Marco Mondelli. “Towards Understanding the Word Sensitivity of Attention Layers: A Study via Random Features.” <i>41st International Conference on Machine Learning</i>, vol. 235, ML Research Press, 2024, pp. 4300–28.","ista":"Bombari S, Mondelli M. 2024. Towards understanding the word sensitivity of attention layers: A study via random features. 41st International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 235, 4300–4328."},"corr_author":"1","OA_type":"green"},{"oa":1,"status":"public","date_updated":"2025-01-30T07:46:16Z","main_file_link":[{"open_access":"1","url":"https://openreview.net/forum?id=mXUDDL4r1Q"}],"abstract":[{"lang":"eng","text":"Reinforcement Learning (RL) from temporal logical specifications is a fundamental problem in sequential decision making. One of the basic and core such specification is the reachability specification that requires a target set to be eventually visited. Despite strong empirical results for RL from such specifications, the theoretical guarantees are bleak, including the impossibility of Probably Approximately Correct (PAC) guarantee for reachability specifications. Given the impossibility result, in this work we consider the problem of RL from reachability specifications along with the information of expected conditional distance (ECD). We present (a) lower bound results which establish the necessity of ECD information for PAC guarantees and (b) an algorithm that establishes PAC-guarantees given the ECD information. To the best of our knowledge, this is the first RL from reachability specifications that does not make any assumptions on the underlying environment to learn policies."}],"quality_controlled":"1","title":"Reinforcement learning from reachability specifications: PAC guarantees with expected conditional distance","publisher":"ML Research Press","type":"conference","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication":"41st International Conference on Machine Learning","day":"29","article_processing_charge":"No","year":"2024","date_created":"2025-01-30T07:45:22Z","_id":"18974","date_published":"2024-07-29T00:00:00Z","OA_place":"publisher","citation":{"short":"J. Svoboda, S. Bansal, K. Chatterjee, in:, 41st International Conference on Machine Learning, ML Research Press, 2024, pp. 47331–47344.","ama":"Svoboda J, Bansal S, Chatterjee K. Reinforcement learning from reachability specifications: PAC guarantees with expected conditional distance. In: <i>41st International Conference on Machine Learning</i>. Vol 235. ML Research Press; 2024:47331-47344.","apa":"Svoboda, J., Bansal, S., &#38; Chatterjee, K. (2024). Reinforcement learning from reachability specifications: PAC guarantees with expected conditional distance. In <i>41st International Conference on Machine Learning</i> (Vol. 235, pp. 47331–47344). Vienna, Austria: ML Research Press.","ista":"Svoboda J, Bansal S, Chatterjee K. 2024. Reinforcement learning from reachability specifications: PAC guarantees with expected conditional distance. 41st International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 235, 47331–47344.","chicago":"Svoboda, Jakub, Suguman Bansal, and Krishnendu Chatterjee. “Reinforcement Learning from Reachability Specifications: PAC Guarantees with Expected Conditional Distance.” In <i>41st International Conference on Machine Learning</i>, 235:47331–44. ML Research Press, 2024.","ieee":"J. Svoboda, S. Bansal, and K. Chatterjee, “Reinforcement learning from reachability specifications: PAC guarantees with expected conditional distance,” in <i>41st International Conference on Machine Learning</i>, Vienna, Austria, 2024, vol. 235, pp. 47331–47344.","mla":"Svoboda, Jakub, et al. “Reinforcement Learning from Reachability Specifications: PAC Guarantees with Expected Conditional Distance.” <i>41st International Conference on Machine Learning</i>, vol. 235, ML Research Press, 2024, pp. 47331–44."},"corr_author":"1","OA_type":"green","alternative_title":["PMLR"],"conference":{"end_date":"2024-07-27","start_date":"2024-07-21","location":"Vienna, Austria","name":"ICML: International Conference on Machine Learning"},"department":[{"_id":"KrCh"}],"author":[{"orcid":"0000-0002-1419-3267","last_name":"Svoboda","full_name":"Svoboda, Jakub","first_name":"Jakub","id":"130759D2-D7DD-11E9-87D2-DE0DE6697425"},{"first_name":"Suguman","full_name":"Bansal, Suguman","last_name":"Bansal"},{"orcid":"0000-0002-4561-241X","last_name":"Chatterjee","full_name":"Chatterjee, Krishnendu","id":"2E5DCA20-F248-11E8-B48F-1D18A9856A87","first_name":"Krishnendu"}],"publication_status":"published","volume":235,"month":"07","scopus_import":"1","intvolume":"       235","language":[{"iso":"eng"}],"oa_version":"Preprint","page":"47331-47344"},{"language":[{"iso":"eng"}],"oa_version":"Preprint","page":"35910-35933","scopus_import":"1","intvolume":"       235","acknowledgement":"The authors thank Adrian Vladu, Razvan Pascanu, Alexandra Peste, Mher Safaryan for their valuable feedback, the IT department from Institute of Science and Technology Austria for the hardware support and Weights and Biases for the infrastructure to track all our experiments.","department":[{"_id":"DaAl"}],"conference":{"location":"Vienna, Austria","start_date":"2024-07-21","end_date":"2024-07-27","name":"ICML: International Conference on Machine Learning"},"author":[{"last_name":"Modoranu","full_name":"Modoranu, Ionut-Vlad","id":"449f7a18-f128-11eb-9611-9b430c0c6333","first_name":"Ionut-Vlad"},{"id":"44b7120e-eb97-11eb-a6c2-e1557aa81d02","first_name":"Aleksei","orcid":"0000-0003-2189-3904","last_name":"Kalinov","full_name":"Kalinov, Aleksei"},{"first_name":"Eldar","id":"47beb3a5-07b5-11eb-9b87-b108ec578218","full_name":"Kurtic, Eldar","last_name":"Kurtic"},{"last_name":"Frantar","full_name":"Frantar, Elias","first_name":"Elias","id":"09a8f98d-ec99-11ea-ae11-c063a7b7fe5f"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","first_name":"Dan-Adrian","full_name":"Alistarh, Dan-Adrian","last_name":"Alistarh","orcid":"0000-0003-3650-940X"}],"publication_status":"published","volume":235,"month":"07","alternative_title":["PMLR"],"citation":{"ieee":"I.-V. Modoranu, A. Kalinov, E. Kurtic, E. Frantar, and D.-A. Alistarh, “Error feedback can accurately compress preconditioners,” in <i>41st International Conference on Machine Learning</i>, Vienna, Austria, 2024, vol. 235, pp. 35910–35933.","chicago":"Modoranu, Ionut-Vlad, Aleksei Kalinov, Eldar Kurtic, Elias Frantar, and Dan-Adrian Alistarh. “Error Feedback Can Accurately Compress Preconditioners.” In <i>41st International Conference on Machine Learning</i>, 235:35910–33. ML Research Press, 2024.","mla":"Modoranu, Ionut-Vlad, et al. “Error Feedback Can Accurately Compress Preconditioners.” <i>41st International Conference on Machine Learning</i>, vol. 235, ML Research Press, 2024, pp. 35910–33.","ista":"Modoranu I-V, Kalinov A, Kurtic E, Frantar E, Alistarh D-A. 2024. Error feedback can accurately compress preconditioners. 41st International Conference on Machine Learning. ICML: International Conference on Machine Learning, PMLR, vol. 235, 35910–35933.","ama":"Modoranu I-V, Kalinov A, Kurtic E, Frantar E, Alistarh D-A. Error feedback can accurately compress preconditioners. In: <i>41st International Conference on Machine Learning</i>. Vol 235. ML Research Press; 2024:35910-35933.","apa":"Modoranu, I.-V., Kalinov, A., Kurtic, E., Frantar, E., &#38; Alistarh, D.-A. (2024). Error feedback can accurately compress preconditioners. In <i>41st International Conference on Machine Learning</i> (Vol. 235, pp. 35910–35933). Vienna, Austria: ML Research Press.","short":"I.-V. Modoranu, A. Kalinov, E. Kurtic, E. Frantar, D.-A. Alistarh, in:, 41st International Conference on Machine Learning, ML Research Press, 2024, pp. 35910–35933."},"corr_author":"1","OA_type":"green","OA_place":"repository","date_created":"2025-01-30T07:53:22Z","acknowledged_ssus":[{"_id":"CampIT"}],"_id":"18975","external_id":{"arxiv":["2306.06098"]},"date_published":"2024-07-30T00:00:00Z","year":"2024","day":"30","article_processing_charge":"No","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","publication":"41st International Conference on Machine Learning","abstract":[{"lang":"eng","text":"Leveraging second-order information about the loss at the scale of deep networks is one of the main lines of approach for improving the performance of current optimizers for deep learning. Yet, existing approaches for accurate full-matrix preconditioning, such as Full-Matrix Adagrad (GGT) or Matrix-Free Approximate Curvature (M-FAC) suffer from massive storage costs when applied even to small-scale models, as they must store a sliding window of gradients, whose memory requirements are multiplicative in the model dimension. In this paper, we address this issue via a novel and efficient error-feedback technique that can be applied to compress preconditioners by up to two orders of magnitude in practice, without loss of convergence. Specifically, our approach compresses the gradient information via sparsification or low-rank compression before it is fed into the preconditioner, feeding the compression error back into future iterations. Extensive experiments on deep neural networks show that this approach can compress full-matrix preconditioners to up to 99% sparsity without accuracy loss, effectively removing the memory overhead of fullmatrix preconditioners such as GGT and M-FAC."}],"title":"Error feedback can accurately compress preconditioners","quality_controlled":"1","publisher":"ML Research Press","publication_identifier":{"eissn":["2640-3498"]},"main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2306.06098"}],"arxiv":1,"status":"public","date_updated":"2025-01-30T07:54:16Z","oa":1},{"day":"15","article_processing_charge":"No","publication":"Proceedings of The 27th International Conference on Artificial Intelligence and Statistics","type":"conference","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"18976","external_id":{"arxiv":["2310.20452"]},"date_published":"2024-05-15T00:00:00Z","date_created":"2025-01-30T08:15:49Z","year":"2024","date_updated":"2025-04-14T07:54:52Z","status":"public","project":[{"call_identifier":"H2020","grant_number":"101034413","name":"IST-BRIDGE: International postdoctoral program","_id":"fc2ed2f7-9c52-11eb-aca3-c01059dda49c"}],"arxiv":1,"oa":1,"quality_controlled":"1","abstract":[{"text":"We analyze asynchronous-type algorithms for distributed SGD in the heterogeneous setting, where each worker has its own computation and communication speeds, as well as data distribution. In these algorithms, workers compute possibly stale and stochastic gradients associated with their local data at some iteration back in history and then return those gradients to the server without synchronizing with other workers. We present a unified convergence theory for non-convex smooth functions in the heterogeneous regime. The proposed analysis provides convergence for pure asynchronous SGD and its various modifications. Moreover, our theory explains what affects the convergence rate and what can be done to improve the performance of asynchronous algorithms. In particular, we introduce a novel asynchronous method based on worker shuffling. As a by-product of our analysis, we also demonstrate convergence guarantees for gradient-type algorithms such as SGD with random reshuffling and shuffle-once mini-batch SGD. The derived rates match the best-known results for those algorithms, highlighting the tightness of our approach. Finally, our numerical evaluations support theoretical findings and show the good practical performance of our method.","lang":"eng"}],"title":"AsGrad: A sharp unified analysis of asynchronous-SGD algorithms","publisher":"ML Research Press","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2310.20452","open_access":"1"}],"publication_identifier":{"eissn":["2640-3498"]},"acknowledgement":"The authors thank all anonymous reviewers for their valuable comments and suggestions on how to improve the manuscript. This work was done when Rustem Islamov was a Master’s student at Institut Polytechnique de Paris (IP Paris) and an intern at Institute of Science and Technology Austria (ISTA). The research of Rustem Islamov was supported by ISTA internship\r\nprogram. Mher Safaryan has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 101034413.","scopus_import":"1","intvolume":"       238","volume":238,"month":"05","author":[{"last_name":"Islamov","full_name":"Islamov, Rustem","first_name":"Rustem"},{"id":"dd546b39-0804-11ed-9c55-ef075c39778d","first_name":"Mher","full_name":"Safaryan, Mher","last_name":"Safaryan"},{"orcid":"0000-0003-3650-940X","full_name":"Alistarh, Dan-Adrian","last_name":"Alistarh","first_name":"Dan-Adrian","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87"}],"conference":{"name":"AISTATS: Conference on Artificial Intelligence and Statistics","end_date":"2024-05-04","start_date":"2024-05-02","location":"Valencia, Spain"},"department":[{"_id":"DaAl"}],"publication_status":"published","oa_version":"Preprint","page":"649-657","language":[{"iso":"eng"}],"OA_place":"repository","alternative_title":["PMLR"],"ec_funded":1,"OA_type":"green","citation":{"short":"R. Islamov, M. Safaryan, D.-A. Alistarh, in:, Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, ML Research Press, 2024, pp. 649–657.","ama":"Islamov R, Safaryan M, Alistarh D-A. AsGrad: A sharp unified analysis of asynchronous-SGD algorithms. In: <i>Proceedings of The 27th International Conference on Artificial Intelligence and Statistics</i>. Vol 238. ML Research Press; 2024:649-657.","apa":"Islamov, R., Safaryan, M., &#38; Alistarh, D.-A. (2024). AsGrad: A sharp unified analysis of asynchronous-SGD algorithms. In <i>Proceedings of The 27th International Conference on Artificial Intelligence and Statistics</i> (Vol. 238, pp. 649–657). Valencia, Spain: ML Research Press.","ista":"Islamov R, Safaryan M, Alistarh D-A. 2024. AsGrad: A sharp unified analysis of asynchronous-SGD algorithms. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics. AISTATS: Conference on Artificial Intelligence and Statistics, PMLR, vol. 238, 649–657.","chicago":"Islamov, Rustem, Mher Safaryan, and Dan-Adrian Alistarh. “AsGrad: A Sharp Unified Analysis of Asynchronous-SGD Algorithms.” In <i>Proceedings of The 27th International Conference on Artificial Intelligence and Statistics</i>, 238:649–57. ML Research Press, 2024.","ieee":"R. Islamov, M. Safaryan, and D.-A. Alistarh, “AsGrad: A sharp unified analysis of asynchronous-SGD algorithms,” in <i>Proceedings of The 27th International Conference on Artificial Intelligence and Statistics</i>, Valencia, Spain, 2024, vol. 238, pp. 649–657.","mla":"Islamov, Rustem, et al. “AsGrad: A Sharp Unified Analysis of Asynchronous-SGD Algorithms.” <i>Proceedings of The 27th International Conference on Artificial Intelligence and Statistics</i>, vol. 238, ML Research Press, 2024, pp. 649–57."},"corr_author":"1"},{"OA_place":"repository","OA_type":"green","citation":{"ista":"Dettmers T, Svirschevski RA, Egiazarian V, Kuznedelev D, Frantar E, Ashkboos S, Borzunov A, Hoefler T, Alistarh D-A. 2024. SpQR: A sparse-quantized representation for near-lossless LLM weight compression. 12th International Conference on Learning Representations. ICLR: International Conference on Learning Representations.","chicago":"Dettmers, Tim, Ruslan A. Svirschevski, Vage Egiazarian, Denis Kuznedelev, Elias Frantar, Saleh Ashkboos, Alexander Borzunov, Torsten Hoefler, and Dan-Adrian Alistarh. “SpQR: A Sparse-Quantized Representation for near-Lossless LLM Weight Compression.” In <i>12th International Conference on Learning Representations</i>. OpenReview, 2024.","ieee":"T. Dettmers <i>et al.</i>, “SpQR: A sparse-quantized representation for near-lossless LLM weight compression,” in <i>12th International Conference on Learning Representations</i>, Vienna, Austria, 2024.","mla":"Dettmers, Tim, et al. “SpQR: A Sparse-Quantized Representation for near-Lossless LLM Weight Compression.” <i>12th International Conference on Learning Representations</i>, OpenReview, 2024.","short":"T. Dettmers, R.A. Svirschevski, V. Egiazarian, D. Kuznedelev, E. Frantar, S. Ashkboos, A. Borzunov, T. Hoefler, D.-A. Alistarh, in:, 12th International Conference on Learning Representations, OpenReview, 2024.","ama":"Dettmers T, Svirschevski RA, Egiazarian V, et al. SpQR: A sparse-quantized representation for near-lossless LLM weight compression. In: <i>12th International Conference on Learning Representations</i>. OpenReview; 2024.","apa":"Dettmers, T., Svirschevski, R. A., Egiazarian, V., Kuznedelev, D., Frantar, E., Ashkboos, S., … Alistarh, D.-A. (2024). SpQR: A sparse-quantized representation for near-lossless LLM weight compression. In <i>12th International Conference on Learning Representations</i>. Vienna, Austria: OpenReview."},"month":"05","department":[{"_id":"DaAl"}],"conference":{"end_date":"2024-05-11","location":"Vienna, Austria","start_date":"2024-05-07","name":"ICLR: International Conference on Learning Representations"},"author":[{"first_name":"Tim","full_name":"Dettmers, Tim","last_name":"Dettmers"},{"first_name":"Ruslan A.","full_name":"Svirschevski, Ruslan A.","last_name":"Svirschevski"},{"first_name":"Vage","full_name":"Egiazarian, Vage","last_name":"Egiazarian"},{"full_name":"Kuznedelev, Denis","last_name":"Kuznedelev","first_name":"Denis"},{"last_name":"Frantar","full_name":"Frantar, Elias","id":"09a8f98d-ec99-11ea-ae11-c063a7b7fe5f","first_name":"Elias"},{"full_name":"Ashkboos, Saleh","last_name":"Ashkboos","first_name":"Saleh"},{"last_name":"Borzunov","full_name":"Borzunov, Alexander","first_name":"Alexander"},{"first_name":"Torsten","full_name":"Hoefler, Torsten","last_name":"Hoefler"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","first_name":"Dan-Adrian","orcid":"0000-0003-3650-940X","last_name":"Alistarh","full_name":"Alistarh, Dan-Adrian"}],"publication_status":"published","acknowledgement":"Denis Kuznedelev acknowledges the support from the Russian Ministry of Science and Higher\r\nEducation, grant No. 075-10-2021-068. Ruslan Svirschevski and Vage Egiazarian and Denis\r\nKuznedelev were supported by the grant for research centers in the field of AI provided by the\r\nAnalytical Center for the Government of the Russian Federation (ACRF) in accordance with the\r\nagreement on the provision of subsidies (identifier of the agreement 000000D730321P5Q0002) and the agreement with HSE University No. 70-2021-00139.","scopus_import":"1","oa_version":"Preprint","language":[{"iso":"eng"}],"oa":1,"date_updated":"2025-01-30T08:27:47Z","status":"public","arxiv":1,"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2306.03078","open_access":"1"}],"abstract":[{"lang":"eng","text":"Recent advances in large language model (LLM) pretraining have led to high-quality LLMs with impressive abilities. By compressing such LLMs via quantization to 3-4 bits per parameter, they can fit into memory-limited devices such as laptops and mobile phones, enabling personalized use. Quantizing models to 3-4 bits per parameter can lead to moderate to high accuracy losses, especially for smaller models (1-10B parameters), which are suitable for edge deployment. To address this accuracy issue, we introduce the Sparse-Quantized Representation (SpQR), a new compressed format and quantization technique that enables for the first time \\emph{near-lossless} compression of LLMs across model scales while reaching similar compression levels to previous methods. SpQR works by identifying and isolating \\emph{outlier weights}, which cause particularly large quantization errors, and storing them in higher precision while compressing all other weights to 3-4 bits, and achieves relative accuracy losses of less than \r\n in perplexity for highly-accurate LLaMA and Falcon LLMs. This makes it possible to run a 33B parameter LLM on a single 24 GB consumer GPU without performance degradation at 15% speedup, thus making powerful LLMs available to consumers without any downsides. SpQR comes with efficient algorithms for both encoding weights into its format, as well as decoding them efficiently at runtime. Specifically, we provide an efficient GPU inference algorithm for SpQR, which yields faster inference than 16-bit baselines at similar accuracy while enabling memory compression gains of more than 4x."}],"title":"SpQR: A sparse-quantized representation for near-lossless LLM weight compression","quality_controlled":"1","publisher":"OpenReview","publication":"12th International Conference on Learning Representations","type":"conference","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","day":"15","article_processing_charge":"No","year":"2024","external_id":{"arxiv":["2306.03078"]},"_id":"18977","date_published":"2024-05-15T00:00:00Z","date_created":"2025-01-30T08:26:59Z"},{"oa_version":"Preprint","language":[{"iso":"eng"}],"acknowledgement":"This project has received funding from the European Research Council (ERC) under the European\r\nUnion’s Horizon 2020 research and innovation programme, grant no. 788183, from the Wittgenstein Prize,\r\nAustrian Science Fund (FWF), grant no. Z 342-N31, and from the DFG Collaborative Research Center TRR\r\n109, ‘Discretization in Geometry and Dynamics’, Austrian Science Fund (FWF), grant no. I 02979-N35.","month":"06","publication_status":"draft","author":[{"full_name":"Brown, Adam","last_name":"Brown","first_name":"Adam"},{"id":"2B23F01E-F248-11E8-B48F-1D18A9856A87","first_name":"Ondrej","orcid":"0000-0003-0464-3823","last_name":"Draganov","full_name":"Draganov, Ondrej"}],"department":[{"_id":"HeEd"}],"ec_funded":1,"related_material":{"record":[{"relation":"later_version","id":"20323","status":"public"},{"status":"public","id":"18979","relation":"dissertation_contains"}]},"corr_author":"1","citation":{"apa":"Brown, A., &#38; Draganov, O. (n.d.). Discrete microlocal Morse theory. <i>arXiv</i>. <a href=\"https://doi.org/10.48550/arXiv.2209.14993\">https://doi.org/10.48550/arXiv.2209.14993</a>","ama":"Brown A, Draganov O. Discrete microlocal Morse theory. <i>arXiv</i>. doi:<a href=\"https://doi.org/10.48550/arXiv.2209.14993\">10.48550/arXiv.2209.14993</a>","short":"A. Brown, O. Draganov, ArXiv (n.d.).","mla":"Brown, Adam, and Ondrej Draganov. “Discrete Microlocal Morse Theory.” <i>ArXiv</i>, doi:<a href=\"https://doi.org/10.48550/arXiv.2209.14993\">10.48550/arXiv.2209.14993</a>.","chicago":"Brown, Adam, and Ondrej Draganov. “Discrete Microlocal Morse Theory.” <i>ArXiv</i>, n.d. <a href=\"https://doi.org/10.48550/arXiv.2209.14993\">https://doi.org/10.48550/arXiv.2209.14993</a>.","ieee":"A. Brown and O. Draganov, “Discrete microlocal Morse theory,” <i>arXiv</i>. .","ista":"Brown A, Draganov O. Discrete microlocal Morse theory. arXiv, <a href=\"https://doi.org/10.48550/arXiv.2209.14993\">10.48550/arXiv.2209.14993</a>."},"OA_place":"repository","date_published":"2024-06-09T00:00:00Z","_id":"18981","external_id":{"arxiv":["2209.14993"]},"date_created":"2025-01-31T17:03:04Z","year":"2024","article_processing_charge":"No","day":"09","publication":"arXiv","type":"preprint","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"text":"We establish several results combining discrete Morse theory and microlocal sheaf theory in the setting of finite posets and simplicial complexes. Our primary tool is a computationally tractable description of the bounded derived category of sheaves on a poset with the Alexandrov topology. We prove that each bounded complex of sheaves on a finite poset admits a unique (up to isomorphism of complexes) minimal injective resolution, and we provide algorithms for computing minimal injective resolution of an injective complex, as well as several useful functors between derived categories of sheaves. For the constant sheaf on a simplicial complex, we give asymptotically tight bounds on the complexity of computing the minimal injective resolution using those algorithms. Our main result is a novel definition of the discrete microsupport of a bounded complex of sheaves on a finite poset. We detail several foundational properties of the discrete microsupport, as well as a microlocal generalization of the discrete homological Morse theorem and Morse inequalities.","lang":"eng"}],"title":"Discrete microlocal Morse theory","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2209.14993"}],"project":[{"_id":"266A2E9E-B435-11E9-9278-68D0E5697425","name":"Alpha Shape Theory Extended","grant_number":"788183","call_identifier":"H2020"},{"_id":"268116B8-B435-11E9-9278-68D0E5697425","grant_number":"Z00342","name":"Mathematics, Computer Science","call_identifier":"FWF"},{"call_identifier":"FWF","_id":"2561EBF4-B435-11E9-9278-68D0E5697425","grant_number":"I02979-N35","name":"Persistence and stability of geometric complexes"}],"status":"public","doi":"10.48550/arXiv.2209.14993","date_updated":"2026-04-07T11:47:29Z","arxiv":1,"oa":1,"tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"}},{"status":"public","date_updated":"2025-07-07T13:23:49Z","arxiv":1,"oa":1,"tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"file":[{"date_updated":"2025-02-04T13:09:08Z","content_type":"application/pdf","checksum":"75c3091e70bd2916cd94afbf40a0c425","file_id":"18997","success":1,"access_level":"open_access","file_size":5659119,"creator":"dernst","date_created":"2025-02-04T13:09:08Z","relation":"main_file","file_name":"2024_NeurIPS_Chen.pdf"}],"quality_controlled":"1","title":"Identifying general mechanism shifts in linear causal representations","abstract":[{"lang":"eng","text":"We consider the linear causal representation learning setting where we observe a linear mixing of d unknown latent factors, which follow a linear structural causal model. Recent work has shown that it is possible to recover the latent factors as well as the underlying structural causal model over them, up to permutation and scaling, provided that we have at least d environments, each of which corresponds to perfect interventions on a single latent node (factor). After this powerful result, a key open problem faced by the community has been to relax these conditions: allow for coarser than perfect single-node interventions, and allow for fewer than d of them, since the number of latent factors d could be very large. In this work, we consider precisely such a setting, where we allow a smaller than d number of environments, and also allow for very coarse interventions that can very coarsely \\textit{change the entire causal graph over the latent factors}. On the flip side, we relax what we wish to extract to simply the \\textit{list of nodes that have shifted between one or more environments}. We provide a surprising identifiability result that it is indeed possible, under some very mild standard assumptions, to identify the set of shifted nodes. Our identifiability proof moreover is a constructive one: we explicitly provide necessary and sufficient conditions for a node to be a shifted node, and show that we can check these conditions given observed data. Our algorithm lends itself very naturally to the sample setting where instead of just interventional distributions, we are provided datasets of samples from each of these distributions. We corroborate our results on both synthetic experiments as well as an interesting psychometric dataset. The code can be found at https://github.com/TianyuCodings/iLCS."}],"publisher":"Neural Information Processing Systems Foundation","publication_identifier":{"eissn":["1049-5258"]},"day":"25","article_processing_charge":"No","publication":"38th Conference on Neural Information Processing Systems","type":"conference","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","_id":"18996","external_id":{"arxiv":["2410.24059"]},"date_published":"2024-09-25T00:00:00Z","date_created":"2025-02-04T13:09:34Z","file_date_updated":"2025-02-04T13:09:08Z","year":"2024","OA_place":"repository","alternative_title":["Advances in Neural Information Processing Systems"],"has_accepted_license":"1","OA_type":"green","citation":{"apa":"Chen, T., Bello, K., Locatello, F., Aragam, B., &#38; Ravikumar, P. K. (2024). Identifying general mechanism shifts in linear causal representations. In <i>38th Conference on Neural Information Processing Systems</i> (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ama":"Chen T, Bello K, Locatello F, Aragam B, Ravikumar PK. Identifying general mechanism shifts in linear causal representations. In: <i>38th Conference on Neural Information Processing Systems</i>. Vol 37. Neural Information Processing Systems Foundation; 2024.","short":"T. Chen, K. Bello, F. Locatello, B. Aragam, P.K. Ravikumar, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","mla":"Chen, Tianyu, et al. “Identifying General Mechanism Shifts in Linear Causal Representations.” <i>38th Conference on Neural Information Processing Systems</i>, vol. 37, Neural Information Processing Systems Foundation, 2024.","ieee":"T. Chen, K. Bello, F. Locatello, B. Aragam, and P. K. Ravikumar, “Identifying general mechanism shifts in linear causal representations,” in <i>38th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2024, vol. 37.","chicago":"Chen, Tianyu, Kevin Bello, Francesco Locatello, Bryon Aragam, and Pradeep Kumar Ravikumar. “Identifying General Mechanism Shifts in Linear Causal Representations.” In <i>38th Conference on Neural Information Processing Systems</i>, Vol. 37. Neural Information Processing Systems Foundation, 2024.","ista":"Chen T, Bello K, Locatello F, Aragam B, Ravikumar PK. 2024. Identifying general mechanism shifts in linear causal representations. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37."},"scopus_import":"1","ddc":["000"],"intvolume":"        37","volume":37,"month":"09","conference":{"end_date":"2024-12-16","location":"Vancouver, Canada","start_date":"2024-12-16","name":"NeurIPS: Neural Information Processing Systems"},"author":[{"first_name":"Tianyu","full_name":"Chen, Tianyu","last_name":"Chen"},{"last_name":"Bello","full_name":"Bello, Kevin","first_name":"Kevin"},{"id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco","orcid":"0000-0002-4850-0683","last_name":"Locatello","full_name":"Locatello, Francesco"},{"first_name":"Bryon","last_name":"Aragam","full_name":"Aragam, Bryon"},{"full_name":"Ravikumar, Pradeep Kumar","last_name":"Ravikumar","first_name":"Pradeep Kumar"}],"department":[{"_id":"FrLo"}],"publication_status":"published","oa_version":"Published Version","language":[{"iso":"eng"}]},{"publication":"Findings of the Association for Computational Linguistics: EMNLP 2024","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","day":"01","article_processing_charge":"No","file_date_updated":"2025-02-10T08:20:34Z","year":"2024","external_id":{"arxiv":["2404.00500"]},"_id":"18998","date_published":"2024-11-01T00:00:00Z","date_created":"2025-02-04T16:19:28Z","oa":1,"tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"doi":"10.18653/v1/2024.findings-emnlp.705","date_updated":"2025-02-10T08:21:37Z","status":"public","arxiv":1,"file":[{"file_name":"2024_EMNLP_Draganov.pdf","relation":"main_file","date_created":"2025-02-10T08:20:34Z","creator":"dernst","file_size":1312638,"access_level":"open_access","success":1,"file_id":"19016","checksum":"f4416a5962194f0181ab0dc7f9ef93c0","content_type":"application/pdf","date_updated":"2025-02-10T08:20:34Z"}],"quality_controlled":"1","title":"The shape of word embeddings: Quantifying non-isometry with topological data analysis","abstract":[{"lang":"eng","text":"Word embeddings represent language vocabularies as clouds of d-dimensional points. We investigate how information is conveyed by the general shape of these clouds, instead of representing the semantic meaning of each token. Specifically, we use the notion of persistent homology from topological data analysis (TDA) to measure the distances between language pairs from the shape of their unlabeled embeddings. These distances quantify the degree of non-isometry of the embeddings. To distinguish whether these differences are random training errors or capture real information about the languages, we use the computed distance matrices to construct language phylogenetic trees over 81 Indo-European languages. Careful evaluation shows that our reconstructed trees exhibit strong and statistically-significant similarities to the reference."}],"publisher":"Association for Computational Linguistics","month":"11","author":[{"id":"2B23F01E-F248-11E8-B48F-1D18A9856A87","first_name":"Ondrej","full_name":"Draganov, Ondrej","last_name":"Draganov","orcid":"0000-0003-0464-3823"},{"full_name":"Skiena, Steven","last_name":"Skiena","first_name":"Steven"}],"department":[{"_id":"GradSch"},{"_id":"HeEd"}],"conference":{"start_date":"2024-11-12","location":"Miami, FL, United States","end_date":"2024-11-16","name":"EMNLP: Conference on Empirical Methods in Natural Language Processing"},"publication_status":"published","scopus_import":"1","ddc":["500"],"oa_version":"Published Version","page":"12080-12099","language":[{"iso":"eng"}],"OA_place":"publisher","OA_type":"gold","citation":{"apa":"Draganov, O., &#38; Skiena, S. (2024). The shape of word embeddings: Quantifying non-isometry with topological data analysis. In <i>Findings of the Association for Computational Linguistics: EMNLP 2024</i> (pp. 12080–12099). Miami, FL, United States: Association for Computational Linguistics. <a href=\"https://doi.org/10.18653/v1/2024.findings-emnlp.705\">https://doi.org/10.18653/v1/2024.findings-emnlp.705</a>","ama":"Draganov O, Skiena S. The shape of word embeddings: Quantifying non-isometry with topological data analysis. In: <i>Findings of the Association for Computational Linguistics: EMNLP 2024</i>. Association for Computational Linguistics; 2024:12080-12099. doi:<a href=\"https://doi.org/10.18653/v1/2024.findings-emnlp.705\">10.18653/v1/2024.findings-emnlp.705</a>","short":"O. Draganov, S. Skiena, in:, Findings of the Association for Computational Linguistics: EMNLP 2024, Association for Computational Linguistics, 2024, pp. 12080–12099.","mla":"Draganov, Ondrej, and Steven Skiena. “The Shape of Word Embeddings: Quantifying Non-Isometry with Topological Data Analysis.” <i>Findings of the Association for Computational Linguistics: EMNLP 2024</i>, Association for Computational Linguistics, 2024, pp. 12080–99, doi:<a href=\"https://doi.org/10.18653/v1/2024.findings-emnlp.705\">10.18653/v1/2024.findings-emnlp.705</a>.","chicago":"Draganov, Ondrej, and Steven Skiena. “The Shape of Word Embeddings: Quantifying Non-Isometry with Topological Data Analysis.” In <i>Findings of the Association for Computational Linguistics: EMNLP 2024</i>, 12080–99. Association for Computational Linguistics, 2024. <a href=\"https://doi.org/10.18653/v1/2024.findings-emnlp.705\">https://doi.org/10.18653/v1/2024.findings-emnlp.705</a>.","ieee":"O. Draganov and S. Skiena, “The shape of word embeddings: Quantifying non-isometry with topological data analysis,” in <i>Findings of the Association for Computational Linguistics: EMNLP 2024</i>, Miami, FL, United States, 2024, pp. 12080–12099.","ista":"Draganov O, Skiena S. 2024. The shape of word embeddings: Quantifying non-isometry with topological data analysis. Findings of the Association for Computational Linguistics: EMNLP 2024. EMNLP: Conference on Empirical Methods in Natural Language Processing, 12080–12099."},"corr_author":"1","has_accepted_license":"1"},{"external_id":{"arxiv":["2406.04102"]},"_id":"18999","date_published":"2024-06-06T00:00:00Z","date_created":"2025-02-04T16:21:21Z","year":"2024","day":"06","article_processing_charge":"No","publication":"arXiv","type":"preprint","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","abstract":[{"text":"Exploring the shape of point configurations has been a key driver in the evolution of TDA (short for topological data analysis) since its infancy. This survey illustrates the recent efforts to broaden these ideas to model spatial interactions among multiple configurations, each distinguished by a color. It describes advances in this area and prepares the ground for further exploration by mentioning unresolved questions and promising research avenues while focusing on the overlap with discrete geometry.","lang":"eng"}],"title":"Chromatic topological data analysis","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2406.04102","open_access":"1"}],"doi":"10.48550/ARXIV.2406.04102","date_updated":"2025-02-10T08:14:27Z","status":"public","arxiv":1,"oa":1,"tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"oa_version":"Preprint","language":[{"iso":"eng"}],"ddc":["510"],"month":"06","department":[{"_id":"GradSch"},{"_id":"HeEd"}],"author":[{"id":"34D2A09C-F248-11E8-B48F-1D18A9856A87","first_name":"Sebastiano","full_name":"Cultrera di Montesano, Sebastiano","last_name":"Cultrera di Montesano","orcid":"0000-0001-6249-0832"},{"id":"2B23F01E-F248-11E8-B48F-1D18A9856A87","first_name":"Ondrej","last_name":"Draganov","full_name":"Draganov, Ondrej","orcid":"0000-0003-0464-3823"},{"orcid":"0000-0002-9823-6833","full_name":"Edelsbrunner, Herbert","last_name":"Edelsbrunner","first_name":"Herbert","id":"3FB178DA-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Morteza","id":"f86f7148-b140-11ec-9577-95435b8df824","last_name":"Saghafian","full_name":"Saghafian, Morteza"}],"publication_status":"submitted","has_accepted_license":"1","OA_type":"green","citation":{"ieee":"S. Cultrera di Montesano, O. Draganov, H. Edelsbrunner, and M. Saghafian, “Chromatic topological data analysis,” <i>arXiv</i>. .","chicago":"Cultrera di Montesano, Sebastiano, Ondrej Draganov, Herbert Edelsbrunner, and Morteza Saghafian. “Chromatic Topological Data Analysis.” <i>ArXiv</i>, n.d. <a href=\"https://doi.org/10.48550/ARXIV.2406.04102\">https://doi.org/10.48550/ARXIV.2406.04102</a>.","mla":"Cultrera di Montesano, Sebastiano, et al. “Chromatic Topological Data Analysis.” <i>ArXiv</i>, 2406.04102, doi:<a href=\"https://doi.org/10.48550/ARXIV.2406.04102\">10.48550/ARXIV.2406.04102</a>.","ista":"Cultrera di Montesano S, Draganov O, Edelsbrunner H, Saghafian M. Chromatic topological data analysis. arXiv, 2406.04102.","ama":"Cultrera di Montesano S, Draganov O, Edelsbrunner H, Saghafian M. Chromatic topological data analysis. <i>arXiv</i>. doi:<a href=\"https://doi.org/10.48550/ARXIV.2406.04102\">10.48550/ARXIV.2406.04102</a>","apa":"Cultrera di Montesano, S., Draganov, O., Edelsbrunner, H., &#38; Saghafian, M. (n.d.). Chromatic topological data analysis. <i>arXiv</i>. <a href=\"https://doi.org/10.48550/ARXIV.2406.04102\">https://doi.org/10.48550/ARXIV.2406.04102</a>","short":"S. Cultrera di Montesano, O. Draganov, H. Edelsbrunner, M. Saghafian, ArXiv (n.d.)."},"corr_author":"1","article_number":"2406.04102","OA_place":"repository"},{"publication":"38th Conference on Neural Information Processing Systems","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","type":"conference","day":"01","article_processing_charge":"No","file_date_updated":"2025-02-05T07:44:58Z","year":"2024","_id":"19005","external_id":{"arxiv":["2405.13888"]},"date_published":"2024-12-01T00:00:00Z","date_created":"2025-02-05T07:49:00Z","oa":1,"tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)"},"date_updated":"2025-07-10T11:51:32Z","status":"public","arxiv":1,"file":[{"checksum":"fe8832367e7143876f178244385d859e","file_id":"19006","content_type":"application/pdf","success":1,"date_updated":"2025-02-05T07:44:58Z","file_name":"2024_NeurIPS_Yao.pdf","file_size":2595855,"creator":"dernst","access_level":"open_access","date_created":"2025-02-05T07:44:58Z","relation":"main_file"}],"abstract":[{"text":"Causal representation learning promises to extend causal models to hidden causal\r\nvariables from raw entangled measurements. However, most progress has focused\r\non proving identifiability results in different settings, and we are not aware of any\r\nsuccessful real-world application. At the same time, the field of dynamical systems\r\nbenefited from deep learning and scaled to countless applications but does not allow\r\nparameter identification. In this paper, we draw a clear connection between the two\r\nand their key assumptions, allowing us to apply identifiable methods developed\r\nin causal representation learning to dynamical systems. At the same time, we can\r\nleverage scalable differentiable solvers developed for differential equations to build\r\nmodels that are both identifiable and practical. Overall, we learn explicitly controllable models that isolate the trajectory-specific parameters for further downstream\r\ntasks such as out-of-distribution classification or treatment effect estimation. We\r\nexperiment with a wind simulator with partially known factors of variation. We\r\nalso apply the resulting model to real-world climate data and successfully answer\r\ndownstream causal questions in line with existing literature on climate change.\r\nCode is available at https://github.com/CausalLearningAI/crl-dynamical-systems.","lang":"eng"}],"quality_controlled":"1","title":"Marrying causal representation learning with dynamical systems for science","publisher":"Neural Information Processing Systems Foundation","month":"12","volume":37,"conference":{"name":"NeurIPS: Neural Information Processing Systems","end_date":"2024-12-16","location":"Vancouver, Canada","start_date":"2024-12-16"},"department":[{"_id":"CaMu"},{"_id":"FrLo"}],"author":[{"full_name":"Yao, Dingling","last_name":"Yao","id":"d3e02e50-48a8-11ee-8f62-c108061797fa","first_name":"Dingling"},{"full_name":"Muller, Caroline J","last_name":"Muller","orcid":"0000-0001-5836-5350","first_name":"Caroline J","id":"f978ccb0-3f7f-11eb-b193-b0e2bd13182b"},{"orcid":"0000-0002-4850-0683","last_name":"Locatello","full_name":"Locatello, Francesco","first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4"}],"publication_status":"published","acknowledgement":"We thank Niklas Boers for recommending the SpeedyWeather simulator and Valentino Maiorca\r\nfor guidance on Fourier transformation for SST data. We are also grateful to Shimeng Huang and Riccardo Cadei for their feedback on the treatment effect estimation experiment and to Jiale Chen and Adeel Pervez for their assistance with the solver implementation. Finally, we appreciate the anonymous reviewers for their insightful suggestions, which helped improve the manuscript. ","scopus_import":"1","ddc":["000","550"],"intvolume":"        37","oa_version":"Published Version","language":[{"iso":"eng"}],"OA_place":"publisher","OA_type":"gold","citation":{"ista":"Yao D, Muller CJ, Locatello F. 2024. Marrying causal representation learning with dynamical systems for science. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","ieee":"D. Yao, C. J. Muller, and F. Locatello, “Marrying causal representation learning with dynamical systems for science,” in <i>38th Conference on Neural Information Processing Systems</i>, Vancouver, Canada, 2024, vol. 37.","chicago":"Yao, Dingling, Caroline J Muller, and Francesco Locatello. “Marrying Causal Representation Learning with Dynamical Systems for Science.” In <i>38th Conference on Neural Information Processing Systems</i>, Vol. 37. Neural Information Processing Systems Foundation, 2024.","mla":"Yao, Dingling, et al. “Marrying Causal Representation Learning with Dynamical Systems for Science.” <i>38th Conference on Neural Information Processing Systems</i>, vol. 37, Neural Information Processing Systems Foundation, 2024.","short":"D. Yao, C.J. Muller, F. Locatello, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","ama":"Yao D, Muller CJ, Locatello F. Marrying causal representation learning with dynamical systems for science. In: <i>38th Conference on Neural Information Processing Systems</i>. Vol 37. Neural Information Processing Systems Foundation; 2024.","apa":"Yao, D., Muller, C. J., &#38; Locatello, F. (2024). Marrying causal representation learning with dynamical systems for science. In <i>38th Conference on Neural Information Processing Systems</i> (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation."},"corr_author":"1","related_material":{"link":[{"relation":"software","url":"https://github.com/CausalLearningAI/crl-dynamical-systems"}]},"alternative_title":["Advances in Neural Information Processing Systems"],"has_accepted_license":"1"}]