{"updated":"2025-01-20T06:16:00.644512+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00175361","sets":["6164:6165:6210:8931"]},"path":["8931"],"owner":"11","recid":"175361","title":["線形関数近似によるトリックテイキングゲームのQ学習"],"pubdate":{"attribute_name":"公開日","attribute_value":"2016-10-28"},"_buckets":{"deposit":"b9abea36-3547-48e5-b841-662176d40b87"},"_deposit":{"id":"175361","pid":{"type":"depid","value":"175361","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"線形関数近似によるトリックテイキングゲームのQ学習","author_link":["365148","365150","365149","365151"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"線形関数近似によるトリックテイキングゲームのQ学習"},{"subitem_title":"Q-learning for Trick-Taking Card Games with linear function approximation","subitem_title_language":"en"}]},"item_type_id":"18","publish_date":"2016-10-28","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_18_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京大学工学部電子情報工学科"},{"subitem_text_value":"東京大学大学院工学系研究科電気系工学専攻"}]},"item_18_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Department of Information and Communication Engineering, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Depertment of Electrical Engineering and Information Systems, Graduate School of Engineering, The University of Tokyo","subitem_text_language":"en"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/175361/files/IPSJ-GPWS2016032.pdf","label":"IPSJ-GPWS2016032.pdf"},"date":[{"dateType":"Available","dateValue":"2016-10-28"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-GPWS2016032.pdf","filesize":[{"value":"724.0 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"5"},{"tax":["include_tax"],"price":"0","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"18"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"c3701239-4873-4b53-b38a-f7bd2df20777","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2016 by the Information Processing Society of Japan"}]},"item_18_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"齋藤, 雄太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"鶴岡, 慶雅"}],"nameIdentifiers":[{}]}]},"item_18_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yuta, Saito","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yoshimasa, Tsuruoka","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_18_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"事前知識を用いない多人数不完全情報ゲームのAIの学習は、人工知能を現実世界の問題に応用する上で非常に重要な課題の一つである。本研究では、多人数不完全情報ゲームの一種であるトリックテイキングゲームの行動価値観数を線形関数で近似し、Q学習を行った。その結果、トリックテイキングゲームにQ学習を適用することで単純なルールベースのプレイヤに勝る結果が得られること、自己対戦による学習を行うことで、ランダムプレイヤによる学習を行った時よりも学習結果が向上することを示した。","subitem_description_type":"Other"}]},"item_18_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Learning the AI of a multiplayer imperfect information game without prior knowledge is one of the important challenges toward the application of AI to real-world problems. In this study, we attempted to learn action-value functions for trick-taking games, which is a kind of multiplayer imperfect information games. We built linear action-value functions using Q-learning. Experimental results show that the player built by Q-learning is superior to a simple rule-based player and that learning with self-play is better than using a random player as the opponent.","subitem_description_type":"Other"}]},"item_18_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"200","bibliographic_titles":[{"bibliographic_title":"ゲームプログラミングワークショップ2016論文集"}],"bibliographicPageStart":"196","bibliographicIssueDates":{"bibliographicIssueDate":"2016-10-28","bibliographicIssueDateType":"Issued"},"bibliographicVolumeNumber":"2016"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"created":"2025-01-19T00:45:18.529523+00:00","id":175361,"links":{}}