{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,21]],"date-time":"2026-07-21T15:11:40Z","timestamp":1784646700163,"version":"3.55.0"},"reference-count":69,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","license":[{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["51935005"],"award-info":[{"award-number":["51935005"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Fundamental Research Program","award":["JCKY20200603C010"],"award-info":[{"award-number":["JCKY20200603C010"]}]},{"name":"Science and Technology on Space Intelligent Laboratory","award":["ZDSYS-2018-02"],"award-info":[{"award-number":["ZDSYS-2018-02"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Cybern."],"published-print":{"date-parts":[[2023,1]]},"DOI":"10.1109\/tcyb.2021.3107202","type":"journal-article","created":{"date-parts":[[2021,9,8]],"date-time":"2021-09-08T16:09:37Z","timestamp":1631117377000},"page":"392-405","source":"Crossref","is-referenced-by-count":17,"title":["Addressing Hindsight Bias in Multigoal Reinforcement Learning"],"prefix":"10.1109","volume":"53","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8379-9385","authenticated-orcid":false,"given":"Chenjia","family":"Bai","sequence":"first","affiliation":[{"name":"Department of Computer Science and Technology, Harbin Institute of Technology, Harbin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1654-4681","authenticated-orcid":false,"given":"Lingxiao","family":"Wang","sequence":"additional","affiliation":[{"name":"Departments of Industrial Engineering and Management Sciences, Northwestern University, Evanston, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6617-4842","authenticated-orcid":false,"given":"Yixin","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of EECS, University of California at Berkeley, Berkeley, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhaoran","family":"Wang","sequence":"additional","affiliation":[{"name":"Departments of Industrial Engineering and Management Sciences, Northwestern University, Evanston, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5147-2695","authenticated-orcid":false,"given":"Rui","family":"Zhao","sequence":"additional","affiliation":[{"name":"Tencent AI Lab, Shenzhen, Tencent, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3510-390X","authenticated-orcid":false,"given":"Chenyao","family":"Bai","sequence":"additional","affiliation":[{"name":"Academy for Engineering & Technology, Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6568-1335","authenticated-orcid":false,"given":"Peng","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Harbin Institute of Technology, Harbin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","volume-title":"Reinforcement Learning: An Introduction","author":"Sutton","year":"2018"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11796"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1038\/nature14236"},{"key":"ref4","first-page":"13","article-title":"Model-based reinforcement learning for atari","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Kaiser"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1126\/science.aar6404"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1038\/nature24270"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1177\/0278364917710318"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1177\/0278364919887447"},{"key":"ref9","volume-title":"Multi-goal reinforcement learning: Challenging robotics environments and request for research","author":"Plappert","year":"2018"},{"key":"ref10","first-page":"5048","article-title":"Hindsight experience replay","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Andrychowicz"},{"key":"ref11","first-page":"1057","article-title":"Policy gradient methods for reinforcement learning with function approximation","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Sutton"},{"key":"ref12","first-page":"1785","article-title":"Better exploration with optimistic actor critic","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Ciosek"},{"key":"ref13","first-page":"1587","article-title":"Addressing function approximation error in actor-critic methods","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Fujimoto"},{"key":"ref14","first-page":"387","article-title":"Deterministic policy gradient algorithms","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","volume":"32","author":"Silver"},{"key":"ref15","first-page":"13","article-title":"Continuous control with deep reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Lillicrap"},{"key":"ref16","first-page":"278","article-title":"Policy invariance under reward transformations: Theory and application to reward shaping","volume-title":"Proc. 32nd Int. Conf. Mach. Learn. (ICML)","author":"Ng"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1201\/9780203711606"},{"key":"ref18","first-page":"13","article-title":"On bonus based exploration methods in the arcade learning environment","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Taiga"},{"key":"ref19","first-page":"13","article-title":"Dynamic-aware unsupervised discovery of skills","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Sharma"},{"key":"ref20","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511803161","volume-title":"Causality","author":"Pearl","year":"2009"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1952.10483446"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/70.1.41"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1987.10478441"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2019.2890974"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.3015811"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.2977661"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.3023033"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2014.2313655"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.3037276"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2019.2939174"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2021.3053414"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2014.2352038"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2019.2949596"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.2977374"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.3028378"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2021.3049555"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.3002892"},{"key":"ref38","first-page":"761","article-title":"Horde: A scalable real-time architecture for learning knowledge from unsupervised sensorimotor interaction","volume-title":"Proc. Int. Conf. Auton. Agents Multiagent Syst.","volume":"2","author":"Sutton"},{"key":"ref39","first-page":"1094","article-title":"Learning to achieve goals","volume-title":"Proc. Int. Joint Conf. Artif. Intell. (IJCAI)","author":"Kaelbling"},{"key":"ref40","first-page":"1312","article-title":"Universal value function approximators","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","volume":"37","author":"Schaul"},{"key":"ref41","article-title":"ARCHER: Aggressive rewards to counter bias in hindsight experience replay","author":"Lanka","year":"2018"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.2990722"},{"key":"ref43","first-page":"577","article-title":"Principled exploration via optimistic bootstrapping and backward induction","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Bai"},{"issue":"3","key":"ref44","first-page":"182","article-title":"Survey on sparse reward in deep reinforcement learning","volume":"47","author":"Yang","year":"2020","journal-title":"J. Comput. Sci."},{"key":"ref45","first-page":"13","article-title":"Hindsight policy gradients","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Rauber"},{"key":"ref46","first-page":"9191","article-title":"Visual reinforcement learning with imagined goals","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Nair"},{"key":"ref47","first-page":"13","article-title":"Auto-encoding variational bayes","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Kingma"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1561\/9781680836233"},{"key":"ref49","first-page":"13","article-title":"DHER: Hindsight experience replay for dynamic goals","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Fang"},{"key":"ref50","first-page":"7553","article-title":"Maximum entropy-regularized multi-goal reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Zhao"},{"key":"ref51","first-page":"12602","article-title":"Curriculum-guided hindsight experience replay","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Fang"},{"key":"ref52","first-page":"13464","article-title":"Exploration via hindsight goal generation","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Ren"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2020.106140"},{"key":"ref54","first-page":"13","article-title":"Competitive experience replay","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Liu"},{"key":"ref55","first-page":"13","article-title":"Intrinsic motivation and automatic curricula via asymmetric self-play","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Sukhbaatar"},{"key":"ref56","first-page":"13","article-title":"Learning multi-level hierarchies with hindsight","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Levy"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(99)00052-1"},{"key":"ref58","first-page":"1940","article-title":"Mapping state space using landmarks for universal goal reaching","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Huang"},{"key":"ref59","first-page":"1331","article-title":"Curious: Intrinsically motivated modular multi-goal reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Colas"},{"key":"ref60","first-page":"15298","article-title":"Goal-conditioned imitation learning","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Ding"},{"key":"ref61","first-page":"13","article-title":"CM3: Cooperative multi-goal multi-stage multi-agent reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Yang"},{"key":"ref62","first-page":"14814","article-title":"Planning with goal-conditioned policies","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Nasiriany"},{"key":"ref63","volume-title":"OpenAI gym","author":"Brockman","year":"2016"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v29i1.9486"},{"key":"ref65","volume-title":"Machine Learning: A Probabilistic Perspective","author":"Murphy","year":"2012"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.31390\/gradschool_dissertations.4601"},{"key":"ref67","article-title":"Representation learning with contrastive predictive coding","author":"Oord","year":"2018"},{"key":"ref68","article-title":"Demystifying contrastive self-supervised learning: Invariances, augmentations and dataset biases","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Purushwalkam"},{"key":"ref69","first-page":"13","article-title":"Learning invariant representations for reinforcement learning without reconstruction","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Zhang"}],"container-title":["IEEE Transactions on Cybernetics"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6221036\/9998320\/09531338.pdf?arnumber=9531338","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T18:05:29Z","timestamp":1761588329000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9531338\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1]]},"references-count":69,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/tcyb.2021.3107202","relation":{},"ISSN":["2168-2267","2168-2275"],"issn-type":[{"value":"2168-2267","type":"print"},{"value":"2168-2275","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,1]]}}}