{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T18:55:03Z","timestamp":1763751303305,"version":"3.45.0"},"reference-count":55,"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:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2024YFB2505500"],"award-info":[{"award-number":["2024YFB2505500"]}]},{"name":"Beijing Natural Science Foundation","award":["L257002"],"award-info":[{"award-number":["L257002"]}]},{"DOI":"10.13039\/501100001809","name":"NSFC, China","doi-asserted-by":"publisher","award":["52202487"],"award-info":[{"award-number":["52202487"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Automat. Sci. Eng."],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/tase.2025.3631883","type":"journal-article","created":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T18:44:27Z","timestamp":1762973067000},"page":"24105-24119","source":"Crossref","is-referenced-by-count":0,"title":["Multimodal Reinforcement Learning With Score-Based Policy"],"prefix":"10.1109","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-5331-0938","authenticated-orcid":false,"given":"Wenjun","family":"Zou","sequence":"first","affiliation":[{"name":"School of Vehicle and Mobility, Tsinghua University, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6252-0618","authenticated-orcid":false,"given":"Yinuo","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Vehicle and Mobility, Tsinghua University, Beijing, China"}]},{"given":"Tong","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Vehicle and Mobility, Tsinghua University, Beijing, China"}]},{"given":"Bin","family":"Shuai","sequence":"additional","affiliation":[{"name":"School of Vehicle and Mobility, Tsinghua University, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4566-1103","authenticated-orcid":false,"given":"Liming","family":"Xiao","sequence":"additional","affiliation":[{"name":"School of Mechanical Engineering, University of Science and Technology Beijing, Beijing, China"}]},{"given":"Yinsong","family":"Ma","sequence":"additional","affiliation":[{"name":"Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, MD, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3697-1576","authenticated-orcid":false,"given":"Jingliang","family":"Duan","sequence":"additional","affiliation":[{"name":"School of Mechanical Engineering, University of Science and Technology Beijing, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4923-3633","authenticated-orcid":false,"given":"Shengbo","family":"Eben Li","sequence":"additional","affiliation":[{"name":"School of Vehicle and Mobility, Tsinghua University, Beijing, China"}]}],"member":"263","reference":[{"doi-asserted-by":"publisher","key":"ref1","DOI":"10.1109\/ICRA57147.2024.10610200"},{"key":"ref2","first-page":"1","article-title":"Hybrid internal model: Learning agile legged locomotion with simulated robot response","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Long"},{"doi-asserted-by":"publisher","key":"ref3","DOI":"10.1038\/s41586-023-05732-2"},{"doi-asserted-by":"publisher","key":"ref4","DOI":"10.1109\/TASE.2023.3342419"},{"doi-asserted-by":"publisher","key":"ref5","DOI":"10.1109\/MCI.2024.3364428"},{"doi-asserted-by":"publisher","key":"ref6","DOI":"10.1109\/TASE.2018.2877499"},{"doi-asserted-by":"publisher","key":"ref7","DOI":"10.1038\/s41586-023-06419-4"},{"doi-asserted-by":"publisher","key":"ref8","DOI":"10.1109\/TASE.2023.3292388"},{"doi-asserted-by":"publisher","key":"ref9","DOI":"10.1109\/TASE.2024.3358894"},{"doi-asserted-by":"publisher","key":"ref10","DOI":"10.1109\/TASE.2024.3377810"},{"key":"ref11","first-page":"1","article-title":"Continuous control with deep reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Lillicrap"},{"key":"ref12","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"arXiv:1707.06347"},{"key":"ref13","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"},{"doi-asserted-by":"publisher","key":"ref14","DOI":"10.1109\/TNNLS.2021.3082568"},{"doi-asserted-by":"publisher","key":"ref15","DOI":"10.1109\/TPAMI.2025.3537087"},{"key":"ref16","first-page":"1352","article-title":"Reinforcement learning with deep energy-based policies","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Haarnoja"},{"key":"ref17","first-page":"1","article-title":"Diversity is all you need: Learning skills without a reward function","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Eysenbach"},{"key":"ref18","article-title":"Policy representation via diffusion probability model for reinforcement learning","author":"Yang","year":"2023","journal-title":"arXiv:2305.13122"},{"doi-asserted-by":"publisher","key":"ref19","DOI":"10.52202\/079017-1708"},{"key":"ref20","article-title":"Policy bifurcation in safe reinforcement learning","author":"Zou","year":"2024","journal-title":"arXiv:2403.12847"},{"key":"ref21","article-title":"Latent-variable advantage-weighted policy optimization for offline RL","author":"Chen","year":"2022","journal-title":"arXiv:2203.08949"},{"key":"ref22","first-page":"1","article-title":"Diffusion policies as an expressive policy class for offline reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Wang"},{"doi-asserted-by":"publisher","key":"ref23","DOI":"10.1109\/ICRA.2019.8794131"},{"key":"ref24","article-title":"Offline reinforcement learning with mixture of deterministic policies","author":"Osa","year":"Sep. 2023","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref25","article-title":"Comparing deterministic and soft policy gradients for optimizing Gaussian mixture actors","author":"Dey","year":"Dec. 2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref26","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Ho"},{"key":"ref27","article-title":"Denoising diffusion implicit models","author":"Song","year":"2020","journal-title":"arXiv:2010.02502"},{"key":"ref28","article-title":"Score-based generative modeling through stochastic differential equations","author":"Song","year":"2020","journal-title":"arXiv:2011.13456"},{"doi-asserted-by":"publisher","key":"ref29","DOI":"10.1177\/02783649241273668"},{"key":"ref30","first-page":"22825","article-title":"Contrastive energy prediction for exact energy-guided diffusion sampling in offline reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Lu"},{"key":"ref31","first-page":"1","article-title":"Safe offline reinforcement learning with feasibility-guided diffusion model","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Zheng"},{"doi-asserted-by":"publisher","key":"ref32","DOI":"10.52202\/079017-1717"},{"key":"ref33","first-page":"1","article-title":"DIME: Diffusion-based maximum entropy reinforcement learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Celik"},{"key":"ref34","first-page":"1","article-title":"Learning a diffusion model policy from rewards via Q-score matching","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Psenka"},{"key":"ref35","first-page":"681","article-title":"Bayesian learning via stochastic gradient Langevin dynamics","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Welling"},{"key":"ref36","first-page":"1587","article-title":"Addressing function approximation error in actor-critic methods","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Fujimoto"},{"doi-asserted-by":"publisher","key":"ref37","DOI":"10.1109\/TAC.1972.1100034"},{"key":"ref38","article-title":"Steering your diffusion policy with latent space reinforcement learning","author":"Wagenmaker","year":"2025","journal-title":"arXiv:2506.15799"},{"key":"ref39","article-title":"GenPO: Generative diffusion models meet on-policy reinforcement learning","author":"Ding","year":"2025","journal-title":"arXiv:2505.18763"},{"key":"ref40","article-title":"Flow matching policy gradients","author":"McAllister","year":"2025","journal-title":"arXiv:2507.21053"},{"doi-asserted-by":"publisher","key":"ref41","DOI":"10.1007\/978-981-19-7784-8"},{"doi-asserted-by":"publisher","key":"ref42","DOI":"10.1109\/TNN.1998.712192"},{"doi-asserted-by":"publisher","key":"ref43","DOI":"10.1162\/089976602760128018"},{"key":"ref44","first-page":"1","article-title":"Generative modeling by estimating gradients of the data distribution","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Song"},{"doi-asserted-by":"publisher","key":"ref45","DOI":"10.1143\/JPSJ.65.1604"},{"doi-asserted-by":"publisher","key":"ref46","DOI":"10.1609\/aaai.v30i1.10295"},{"doi-asserted-by":"publisher","key":"ref47","DOI":"10.48550\/arXiv.1812.05905"},{"doi-asserted-by":"publisher","key":"ref48","DOI":"10.2307\/2984875"},{"volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Kingma","article-title":"Adam: A method for stochastic optimization","key":"ref49"},{"doi-asserted-by":"publisher","key":"ref50","DOI":"10.1109\/IROS.2012.6386109"},{"key":"ref51","article-title":"OpenAI gym","author":"Brockman","year":"2016","journal-title":"arXiv:1606.01540"},{"doi-asserted-by":"publisher","key":"ref52","DOI":"10.1016\/j.commtr.2023.100096"},{"key":"ref53","article-title":"IDQL: Implicit Q-learning as an actor-critic method with diffusion policies","author":"Hansen-Estruch","year":"2023","journal-title":"arXiv:2304.10573"},{"doi-asserted-by":"publisher","key":"ref54","DOI":"10.52202\/079017-3136"},{"key":"ref55","article-title":"UMAP: Uniform manifold approximation and projection for dimension reduction","author":"McInnes","year":"2018","journal-title":"arXiv:1802.03426"}],"container-title":["IEEE Transactions on Automation Science and Engineering"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/8856\/10839176\/11242158.pdf?arnumber=11242158","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T18:46:20Z","timestamp":1763750780000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11242158\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":55,"URL":"https:\/\/doi.org\/10.1109\/tase.2025.3631883","relation":{},"ISSN":["1545-5955","1558-3783"],"issn-type":[{"type":"print","value":"1545-5955"},{"type":"electronic","value":"1558-3783"}],"subject":[],"published":{"date-parts":[[2025]]}}}