{"created":"2025-01-19T00:31:46.443692+00:00","updated":"2025-01-20T13:09:46.662700+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00157979","sets":["1164:5305:8612:8613"]},"path":["8613"],"owner":"11","recid":"157979","title":["方策勾配を用いた教師有り学習によるコンピュータ大貧民の方策関数の学習とモンテカルロシミュレーションへの利用"],"pubdate":{"attribute_name":"公開日","attribute_value":"2016-03-01"},"_buckets":{"deposit":"fb7b23d6-cb39-44b3-b5ea-e3927e90c89e"},"_deposit":{"id":"157979","pid":{"type":"depid","value":"157979","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"方策勾配を用いた教師有り学習によるコンピュータ大貧民の方策関数の学習とモンテカルロシミュレーションへの利用","author_link":["300479","300480","300481","300478"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"方策勾配を用いた教師有り学習によるコンピュータ大貧民の方策関数の学習とモンテカルロシミュレーションへの利用"},{"subitem_title":"Supervised learning of policy function based on policy gradients and application to Monte Carlo simulation in Daihinmin","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ゲーム状態の評価および行動価値の推定と機械学習","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2016-03-01","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京大学大学院総合文化研究科"},{"subitem_text_value":"東京大学情報基盤センター"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Arts and Sciences, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Information Technology Center, The University of Tokyo","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"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/157979/files/IPSJ-GI16035010.pdf","label":"IPSJ-GI16035010.pdf"},"date":[{"dateType":"Available","dateValue":"2018-03-01"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-GI16035010.pdf","filesize":[{"value":"755.8 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"18"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"c3780d60-8fec-4adc-9694-c6750e28c283","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2016 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"大渡, 勝己"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"田中, 哲朗"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Katsuki, Ohto","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tetsuro, Tanaka","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11362144","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8736","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"大貧民は多人数不完全情報ゲームに分類され，日本中で広く親しまれているゲームである．大貧民のコンピュータプログラムの大会 （UECda） も行われており，近年ではモンテカルロ法を用いたプログラムが上位を占めている．本研究では，大貧民の知識を用いた方策関数を設計し，そのパラメータを方策勾配を用いた教師有り学習によって過去のプログラムの棋譜から学習した．その結果，公開されている過去のコンピュータ大貧民のプログラムと比較し，モンテカルロ法を使わないプログラム，使うプログラムのいずれとしても過去最高レベルの強さを達成することが出来た．さらに，他プログラムの棋譜を用いずとも，プログラムの自己対戦棋譜からの学習を繰り返すことで，同等の強さまでプログラムを強くすることに成功した．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告ゲーム情報学（GI）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2016-03-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"10","bibliographicVolumeNumber":"2016-GI-35"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"id":157979,"links":{}}