{"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/unifying-pac-and-regret-uniform-pac-bounds","title":"Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning","arxiv_id":"1703.07710","date":"2017-03-22","proceeding":"NeurIPS 2017 12","authors":["Christoph Dann","Tor Lattimore","Emma Brunskill"],"abstract":"Statistical performance bounds for reinforcement learning (RL) algorithms can\nbe critical for high-stakes applications like healthcare. This paper introduces\na new framework for theoretically measuring the performance of such algorithms\ncalled Uniform-PAC, which is a strengthening of the classical Probably\nApproximately Correct (PAC) framework. In contrast to the PAC framework, the\nuniform version may be used to derive high probability regret guarantees and so\nforms a bridge between the two setups that has been missing in the literature.\nWe demonstrate the benefits of the new framework for finite-state episodic MDPs\nwith a new algorithm that is Uniform-PAC and simultaneously achieves optimal\nregret and PAC guarantees except for a factor of the horizon.","url_abs":"http://arxiv.org/abs/1703.07710v3","url_pdf":"http://arxiv.org/pdf/1703.07710v3.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":"unifying-pac-and-regret-uniform-pac-bounds","repo_url":"https://github.com/chrodan/FiniteEpisodicRL.jl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.07710","atlas_url":"https://app.syntology.ai/?focus=1703.07710","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}