{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,17]],"date-time":"2026-04-17T21:16:52Z","timestamp":1776460612207,"version":"3.51.2"},"reference-count":48,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"11","license":[{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100000781","name":"European Research Council","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000781","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Safe data-driven control for human-centric systems","award":["864686"],"award-info":[{"award-number":["864686"]}]},{"DOI":"10.13039\/100000006","name":"ONR","doi-asserted-by":"publisher","award":["N00014-17-1-2622"],"award-info":[{"award-number":["N00014-17-1-2622"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Clark Foundation Distinguished Fellowship"},{"DOI":"10.13039\/100006754","name":"Army Research Laboratory","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006754","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Swiss National Science Foundation as a part of NCCR Automation"},{"DOI":"10.13039\/501100023650","name":"NCCR Catalysis","doi-asserted-by":"publisher","award":["51NF40 225155"],"award-info":[{"award-number":["51NF40 225155"]}],"id":[{"id":"10.13039\/501100023650","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Automat. Contr."],"published-print":{"date-parts":[[2025,11]]},"DOI":"10.1109\/tac.2025.3575565","type":"journal-article","created":{"date-parts":[[2025,6,2]],"date-time":"2025-06-02T14:05:18Z","timestamp":1748873118000},"page":"7414-7429","source":"Crossref","is-referenced-by-count":1,"title":["Risk-Sensitive Safety Filters for Reinforcement Learning With Probabilistic Guarantees"],"prefix":"10.1109","volume":"70","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6263-5608","authenticated-orcid":false,"given":"Armin","family":"Lederer","sequence":"first","affiliation":[{"name":"Learning and Adaptive Systems Group, Institute for Machine Learning, Department of Computer Science, ETH Zurich, Zurich, Switzerland"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5171-122X","authenticated-orcid":false,"given":"Erfaun","family":"Noorani","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering and the Institute for Systems Research (ISR), University of Maryland, College Park, MD, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4955-8561","authenticated-orcid":false,"given":"John S.","family":"Baras","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering and the Institute for Systems Research (ISR), University of Maryland, College Park, MD, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7819-5926","authenticated-orcid":false,"given":"Sandra","family":"Hirche","sequence":"additional","affiliation":[{"name":"Chair of Information-oriented Control (ITR), School of Computation, Information and Technology, Technical University of Munich, Munich, Germany"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1037\/0033-2909.126.2.220"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-control-042920-020211"},{"key":"ref3","article-title":"Challenges of real-world reinforcement learning","volume-title":"Proc. ICML Workshop Real-Life Reinforcement Learn.","author":"Dulac-Arnold","year":"2019"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-control-071723-102940"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/MCS.2023.3291885"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11797"},{"key":"ref7","first-page":"1","article-title":"Risk-constrained reinforcement learning with percentile risk criteria","volume":"18","author":"Chow","year":"2018","journal-title":"J. Mach. Learn. Res."},{"key":"ref8","first-page":"7555","article-title":"Constrained reinforcement learning has zero duality gap","volume":"32","author":"Paternain","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.23919\/ACC50511.2021.9483182"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2021.3049335"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2021.XVII.077"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160600"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2017.XIII.073"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2016.2638961"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2020.109439"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i9.21201"},{"key":"ref17","first-page":"9672","article-title":"Safety guarantees for neural network dynamic systems via stochastic barrier functions","volume":"35","author":"Mazouz","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/IROS45743.2020.9341190"},{"key":"ref19","first-page":"1724","article-title":"Safe nonlinear control using robust neural Lyapunov-barrier functions","volume-title":"Proc. 5th Conf. Robot Learn.","author":"Dawson","year":"2022"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CDC42340.2020.9303785"},{"key":"ref21","article-title":"Conservative safety critics for exploration","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Bharadhwaj","year":"2021"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3070252"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CDC51059.2022.9992470"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.23919\/ACC55779.2023.10156525"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1016\/j.dcn.2017.05.007"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/LCSYS.2021.3086854"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CDC49753.2023.10383524"},{"key":"ref28","volume-title":"Pattern Recognition and Machine Learning","author":"Bishop","year":"2006"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/3206.001.0001"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.5555\/3295222.3295387"},{"key":"ref31","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton","year":"2017"},{"key":"ref32","volume-title":"Nonlinear Systems","author":"Khalil","year":"2002"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1515\/9781400831050"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2006.875041"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/9.286253"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CDC40024.2019.9030241"},{"key":"ref37","first-page":"1861","article-title":"Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Haarnoja","year":"2018"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1016\/j.automatica.2018.03.076"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.4310\/CIS.2006.v6.n3.a1"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1017\/cbo9781107298019"},{"key":"ref41","article-title":"Uniform error and posterior variance bounds for Gaussian process regression with application to safe control","author":"Lederer","year":"2021"},{"key":"ref42","article-title":"Pointwise feasibility of Gaussian process-based safety-critical control under model uncertainty","author":"Castaeda","year":"2021"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2021.3137059"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2012.6386109"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/TCST.2019.2952317"},{"key":"ref46","first-page":"4768","article-title":"RLlib: Abstractions for distributed reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Liang","year":"2018"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-444-53859-8.00003-5"},{"key":"ref48","article-title":"Constrained model-based reinforcement learning with robust cross-entropy method","volume-title":"Proc. ICLR Workshop Secur. Saf. Mach. Learn. Syst.","author":"Liu","year":"2021"}],"container-title":["IEEE Transactions on Automatic Control"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/9\/11218261\/11020782.pdf?arnumber=11020782","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T21:07:04Z","timestamp":1769807224000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11020782\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11]]},"references-count":48,"journal-issue":{"issue":"11"},"URL":"https:\/\/doi.org\/10.1109\/tac.2025.3575565","relation":{},"ISSN":["0018-9286","1558-2523","2334-3303"],"issn-type":[{"value":"0018-9286","type":"print"},{"value":"1558-2523","type":"electronic"},{"value":"2334-3303","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11]]}}}