{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T15:34:05Z","timestamp":1783611245737,"version":"3.55.0"},"reference-count":64,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"name":"JST SPRING, Japan","award":["JPMJSP2136"],"award-info":[{"award-number":["JPMJSP2136"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/access.2025.3535679","type":"journal-article","created":{"date-parts":[[2025,1,28]],"date-time":"2025-01-28T18:46:46Z","timestamp":1738090006000},"page":"35270-35280","source":"Crossref","is-referenced-by-count":1,"title":["Safe Reinforcement Learning via Episodic Control"],"prefix":"10.1109","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0602-7664","authenticated-orcid":false,"given":"Zhuo","family":"Li","sequence":"first","affiliation":[{"name":"Graduate School of Information Science and Electrical Engineering, Kyushu University, Fukuoka, Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9552-0097","authenticated-orcid":false,"given":"Derui","family":"Zhu","sequence":"additional","affiliation":[{"name":"Chair of Cyber Trust, Technical University of Munich, Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1093-1282","authenticated-orcid":false,"given":"Jens","family":"Grossklags","sequence":"additional","affiliation":[{"name":"Chair of Cyber Trust, Technical University of Munich, Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"22","article-title":"Constrained policy optimization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Achiam"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3543846"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1201\/9781315140223"},{"key":"ref4","first-page":"6172","article-title":"Neurosymbolic reinforcement learning with formally verified exploration","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Anderson"},{"key":"ref5","first-page":"908","article-title":"Safe model-based reinforcement learning with stability guarantees","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Berkenkamp"},{"key":"ref6","article-title":"Dota 2 with large scale deep reinforcement learning","author":"OpenAI","year":"2019","journal-title":"arXiv:1912.06680"},{"key":"ref7","article-title":"Model-free episodic control","author":"Blundell","year":"2016","journal-title":"arXiv:1606.04460"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1016\/j.sysconle.2004.08.007"},{"key":"ref9","article-title":"OpenAI gym","author":"Brockman","year":"2016","journal-title":"arXiv:1606.01540"},{"key":"ref10","article-title":"A primal-dual approach to constrained Markov decision processes","author":"Chen","year":"2021","journal-title":"arXiv:2101.10895"},{"issue":"1","key":"ref11","first-page":"6070","article-title":"Risk-constrained reinforcement learning with percentile risk criteria","volume":"18","author":"Chow","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"ref12","article-title":"Sample-efficient reinforcement learning by breaking the replay ratio barrier","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"D\u2019Oro"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3338906.3338954"},{"issue":"1","key":"ref14","first-page":"1437","article-title":"A comprehensive survey on safe reinforcement learning","volume":"16","author":"Garc\u00eda","year":"2015","journal-title":"J. Mach. Learn. Res."},{"key":"ref15","article-title":"Bullet-safety-gym: A framework for constrained reinforcement learning","author":"Gronauer","year":"2022"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3457538"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1080\/23311916.2018.1502242"},{"key":"ref18","first-page":"10567","article-title":"Fast deep reinforcement learning using online adjustments from the past","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Hansen"},{"key":"ref19","first-page":"4380","article-title":"Generalizable episodic memory for deep reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Hu"},{"key":"ref20","article-title":"Safety-gymnasium","author":"Ji","year":"2023","journal-title":"GitHub Repository"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2021.3054625"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1177\/0278364913495721"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2021\/365"},{"key":"ref24","first-page":"3703","article-title":"Batch policy learning under constraints","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Le"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i7.20694"},{"key":"ref26","first-page":"30313","article-title":"Model-based episodic memory induces dynamic hybrid controls","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Le"},{"key":"ref27","first-page":"889","article-title":"Hippocampal contributions to control: The third way","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"20","author":"Lengyel"},{"key":"ref28","article-title":"Neural episodic control with state abstraction","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Li"},{"key":"ref29","article-title":"Continuous control with deep reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Lillicrap"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/337"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TSP.2019.2925601"},{"key":"ref32","first-page":"13644","article-title":"Constrained variational policy optimization for safe reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Liu"},{"key":"ref33","article-title":"Datasets and benchmarks for offline safe reinforcement learning","author":"Liu","year":"2023","journal-title":"arXiv:2306.09303"},{"key":"ref34","article-title":"Offline reinforcement learning with value-based episodic memory","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Ma"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1111\/1468-0262.00296"},{"key":"ref36","first-page":"1928","article-title":"Asynchronous methods for deep reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Mnih"},{"key":"ref37","article-title":"Playing Atari with deep reinforcement learning","author":"Mnih","year":"2013","journal-title":"arXiv:1312.5602"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/IROS47612.2022.9981405"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-13823-7_31"},{"key":"ref40","first-page":"803","article-title":"Lyapunov design for safe reinforcement learning","volume":"3","author":"Perkins","year":"2003","journal-title":"J. Mach. Learn. Res."},{"key":"ref41","article-title":"Model-free episodic control with state aggregation","author":"Pinto","year":"2020","journal-title":"arXiv:2008.09685"},{"key":"ref42","first-page":"2827","article-title":"Neural episodic control","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Pritzel"},{"key":"ref43","first-page":"1889","article-title":"Trust region policy optimization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Schulman"},{"key":"ref44","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"arXiv:1707.06347"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.13140\/RG.2.2.18893.74727"},{"key":"ref46","article-title":"Learning to be safe: Deep RL with a safety critic","author":"Srinivasan","year":"2020","journal-title":"arXiv:2010.14603"},{"key":"ref47","first-page":"9133","article-title":"Responsive safety in reinforcement learning by PID Lagrangian methods","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Stooke"},{"key":"ref48","first-page":"997","article-title":"Safe exploration for optimization with Gaussian processes","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Sui"},{"key":"ref49","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton","year":"2018"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-44719-9_22"},{"key":"ref51","article-title":"Reward constrained policy optimization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Tessler"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2012.6386109"},{"key":"ref53","first-page":"452","article-title":"Teaching multiple tasks to an RL agent using LTL","volume-title":"Proc. 17th Int. Conf. Auto. Agents Multiagent Syst.","author":"Icarte"},{"key":"ref54","article-title":"Projection-based constrained policy optimization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Yang"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i12.17276"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1145\/3477600"},{"key":"ref57","article-title":"Convergent policy optimization for safe reinforcement learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Yu"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/820"},{"key":"ref59","first-page":"15338","article-title":"First order constrained optimization in policy space","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Zhang"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2019.2939174"},{"key":"ref61","article-title":"Episodic multi-agent reinforcement learning with curiosity-driven exploration","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zheng"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-naacl.294"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/ASE51524.2021.9678871"},{"key":"ref64","article-title":"Episodic reinforcement learning with associative memory","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Zhu"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10820123\/10856010.pdf?arnumber=10856010","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,28]],"date-time":"2025-02-28T18:51:53Z","timestamp":1740768713000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10856010\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":64,"URL":"https:\/\/doi.org\/10.1109\/access.2025.3535679","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025]]}}}