{"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/doubly-robust-off-policy-value-evaluation-for","title":"Doubly Robust Off-policy Value Evaluation for Reinforcement Learning","arxiv_id":"1511.03722","date":"2015-11-11","proceeding":null,"authors":["Nan Jiang","Lihong Li"],"abstract":"We study the problem of off-policy value evaluation in reinforcement learning\n(RL), where one aims to estimate the value of a new policy based on data\ncollected by a different policy. This problem is often a critical step when\napplying RL in real-world problems. Despite its importance, existing general\nmethods either have uncontrolled bias or suffer high variance. In this work, we\nextend the doubly robust estimator for bandits to sequential decision-making\nproblems, which gets the best of both worlds: it is guaranteed to be unbiased\nand can have a much lower variance than the popular importance sampling\nestimators. We demonstrate the estimator's accuracy in several benchmark\nproblems, and illustrate its use as a subroutine in safe policy improvement. We\nalso provide theoretical results on the hardness of the problem, and show that\nour estimator can match the lower bound in certain scenarios.","url_abs":"http://arxiv.org/abs/1511.03722v3","url_pdf":"http://arxiv.org/pdf/1511.03722v3.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":"doubly-robust-off-policy-value-evaluation-for","repo_url":"https://github.com/facebookresearch/Horizon","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"doubly-robust-off-policy-value-evaluation-for","repo_url":"https://github.com/facebookresearch/ReAgent","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"sequential-decision-making","task_name":"Sequential Decision Making"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1511.03722","atlas_url":"https://app.syntology.ai/?focus=1511.03722","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}