{"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/the-in-sample-softmax-for-offline","title":"The In-Sample Softmax for Offline Reinforcement Learning","arxiv_id":"2302.14372","date":"2023-02-28","proceeding":null,"authors":["Chenjun Xiao","Han Wang","Yangchen Pan","Adam White","Martha White"],"abstract":"Reinforcement learning (RL) agents can leverage batches of previously collected data to extract a reasonable control policy. An emerging issue in this offline RL setting, however, is that the bootstrapping update underlying many of our methods suffers from insufficient action-coverage: standard max operator may select a maximal action that has not been seen in the dataset. Bootstrapping from these inaccurate values can lead to overestimation and even divergence. There are a growing number of methods that attempt to approximate an \\emph{in-sample} max, that only uses actions well-covered by the dataset. We highlight a simple fact: it is more straightforward to approximate an in-sample \\emph{softmax} using only actions in the dataset. We show that policy iteration based on the in-sample softmax converges, and that for decreasing temperatures it approaches the in-sample max. We derive an In-Sample Actor-Critic (AC), using this in-sample softmax, and show that it is consistently better or comparable to existing offline RL methods, and is also well-suited to fine-tuning.","url_abs":"https://arxiv.org/abs/2302.14372v2","url_pdf":"https://arxiv.org/pdf/2302.14372v2.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":"the-in-sample-softmax-for-offline","repo_url":"https://github.com/hwang-ua/inac_pytorch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"the-in-sample-softmax-for-offline","repo_url":"https://github.com/linhlpv/Offline-and-OfflineOnline-RL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"the-in-sample-softmax-for-offline","repo_url":"https://github.com/typoverflow/OfflineRL-Lib","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"the-in-sample-softmax-for-offline","repo_url":"https://github.com/zzmtsvv/rl_task","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"offline-rl","task_name":"Offline RL"},{"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":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2302.14372","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}