{"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/self-correcting-models-for-model-based","title":"Self-Correcting Models for Model-Based Reinforcement Learning","arxiv_id":"1612.06018","date":"2016-12-19","proceeding":null,"authors":["Erik Talvitie"],"abstract":"When an agent cannot represent a perfectly accurate model of its\nenvironment's dynamics, model-based reinforcement learning (MBRL) can fail\ncatastrophically. Planning involves composing the predictions of the model;\nwhen flawed predictions are composed, even minor errors can compound and render\nthe model useless for planning. Hallucinated Replay (Talvitie 2014) trains the\nmodel to \"correct\" itself when it produces errors, substantially improving MBRL\nwith flawed models. This paper theoretically analyzes this approach,\nilluminates settings in which it is likely to be effective or ineffective, and\npresents a novel error bound, showing that a model's ability to self-correct is\nmore tightly related to MBRL performance than one-step prediction error. These\nresults inspire an MBRL algorithm for deterministic MDPs with performance\nguarantees that are robust to model class limitations.","url_abs":"http://arxiv.org/abs/1612.06018v2","url_pdf":"http://arxiv.org/pdf/1612.06018v2.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":"self-correcting-models-for-model-based","repo_url":"https://github.com/etalvitie/hdaggermc","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"model-based-reinforcement-learning","task_name":"Model-based Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"model","task_name":"model"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.06018","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}