{"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/biasing-mcts-with-features-for-general-games","title":"Biasing MCTS with Features for General Games","arxiv_id":"1903.08942","date":"2019-03-21","proceeding":null,"authors":["Dennis J. N. J. Soemers","Éric Piette","Cameron Browne"],"abstract":"This paper proposes using a linear function approximator, rather than a deep\nneural network (DNN), to bias a Monte Carlo tree search (MCTS) player for\ngeneral games. This is unlikely to match the potential raw playing strength of\nDNNs, but has advantages in terms of generality, interpretability and resources\n(time and hardware) required for training. Features describing local patterns\nare used as inputs. The features are formulated in such a way that they are\neasily interpretable and applicable to a wide range of general games, and might\nencode simple local strategies. We gradually create new features during the\nsame self-play training process used to learn feature weights. We evaluate the\nplaying strength of an MCTS player biased by learnt features against a standard\nupper confidence bounds for trees (UCT) player in multiple different board\ngames, and demonstrate significantly improved playing strength in the majority\nof them after a small number of self-play training games.","url_abs":"http://arxiv.org/abs/1903.08942v1","url_pdf":"http://arxiv.org/pdf/1903.08942v1.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":"biasing-mcts-with-features-for-general-games","repo_url":"https://github.com/Ludeme/LudiiAI","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"board-games","task_name":"Board Games"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}