{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/maximum-a-posteriori-policy-optimisation","title":"Maximum a Posteriori Policy Optimisation","arxiv_id":"1806.06920","date":"2018-06-14","proceeding":"ICLR 2018 1","authors":["Abbas Abdolmaleki","Jost Tobias Springenberg","Yuval Tassa","Remi Munos","Nicolas Heess","Martin Riedmiller"],"abstract":"We introduce a new algorithm for reinforcement learning called Maximum\naposteriori Policy Optimisation (MPO) based on coordinate ascent on a relative\nentropy objective. We show that several existing methods can directly be\nrelated to our derivation. We develop two off-policy algorithms and demonstrate\nthat they are competitive with the state-of-the-art in deep reinforcement\nlearning. In particular, for continuous control, our method outperforms\nexisting methods with respect to sample efficiency, premature convergence and\nrobustness to hyperparameter settings while achieving similar or better final\nperformance.","url_abs":"http://arxiv.org/abs/1806.06920v1","url_pdf":"http://arxiv.org/pdf/1806.06920v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"maximum-a-posteriori-policy-optimisation","repo_url":"https://github.com/acyclics/MPO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"maximum-a-posteriori-policy-optimisation","repo_url":"https://github.com/deepmind/rgb_stacking","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"maximum-a-posteriori-policy-optimisation","repo_url":"https://github.com/MotorCityCobra/C_plusplus_mpo","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"continuous-control","task_name":"continuous-control"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.06920","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}