{"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/thinking-fast-and-slow-with-deep-learning-and","title":"Thinking Fast and Slow with Deep Learning and Tree Search","arxiv_id":"1705.08439","date":"2017-05-23","proceeding":"NeurIPS 2017 12","authors":["Thomas Anthony","Zheng Tian","David Barber"],"abstract":"Sequential decision making problems, such as structured prediction, robotic\ncontrol, and game playing, require a combination of planning policies and\ngeneralisation of those plans. In this paper, we present Expert Iteration\n(ExIt), a novel reinforcement learning algorithm which decomposes the problem\ninto separate planning and generalisation tasks. Planning new policies is\nperformed by tree search, while a deep neural network generalises those plans.\nSubsequently, tree search is improved by using the neural network policy to\nguide search, increasing the strength of new plans. In contrast, standard deep\nReinforcement Learning algorithms rely on a neural network not only to\ngeneralise plans, but to discover them too. We show that ExIt outperforms\nREINFORCE for training a neural network to play the board game Hex, and our\nfinal tree search agent, trained tabula rasa, defeats MoHex 1.0, the most\nrecent Olympiad Champion player to be publicly released.","url_abs":"http://arxiv.org/abs/1705.08439v4","url_pdf":"http://arxiv.org/pdf/1705.08439v4.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":"thinking-fast-and-slow-with-deep-learning-and","repo_url":"https://github.com/coreylowman/ragz","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"thinking-fast-and-slow-with-deep-learning-and","repo_url":"https://github.com/coreylowman/synthesis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"thinking-fast-and-slow-with-deep-learning-and","repo_url":"https://github.com/evangravelle/AI-projects","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"thinking-fast-and-slow-with-deep-learning-and","repo_url":"https://github.com/richemslie/galvanise_zero","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"deep-learning","task_name":"Deep Learning"},{"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":"sequential-decision-making","task_name":"Sequential Decision Making"},{"task_slug":"structured-prediction","task_name":"Structured Prediction"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.08439","atlas_url":"https://app.syntology.ai/?focus=1705.08439","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}