{"id":150579,"created":"2025-01-19T00:25:19.401861+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00150579","sets":["8512:8654:8655:8540:8541"]},"path":["8541"],"owner":"1","recid":"150579","title":["A-023 モンテカルロ碁におけるポテンシャルモデルを利用した枝刈りの可能性(数理モデル化と問題解決(2),A分野:モデル・アルゴリズム・プログラミング)"],"pubdate":{"attribute_name":"公開日","attribute_value":"2011-09-07"},"_buckets":{"deposit":"b43fa7ba-fcce-46fa-b533-c2c84b66399c"},"_deposit":{"id":"150579","pid":{"type":"depid","value":"150579","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"A-023 モンテカルロ碁におけるポテンシャルモデルを利用した枝刈りの可能性(数理モデル化と問題解決(2),A分野:モデル・アルゴリズム・プログラミング)","author_link":["254070","254069","254066","254067","254065","254068"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"A-023 モンテカルロ碁におけるポテンシャルモデルを利用した枝刈りの可能性(数理モデル化と問題解決(2),A分野:モデル・アルゴリズム・プログラミング)"},{"subitem_title":"A-023 A Probability of a Potential Model Pruning in Monte Carlo Go","subitem_title_language":"en"}]},"item_type_id":"26","publish_date":"2011-09-07","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_26_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"琉球大学大学院理工学研究科 総合知能工学専攻"},{"subitem_text_value":"琉球大学工学部情報工学科"},{"subitem_text_value":"琉球大学工学部情報工学科"}]},"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/150579/files/KJ00008742985.pdf"},"date":[{"dateType":"Available","dateValue":"2011-09-07"}],"format":"application/pdf","filename":"KJ00008742985.pdf","filesize":[{"value":"1.5 MB"}],"mimetype":"application/pdf","accessrole":"open_date","version_id":"60105fb3-eb71-46e6-a774-9467a9de71c0","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2011 by IEICE,IPSJ"}]},"item_26_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"大島, 真"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"山田, 孝治"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"遠藤, 聡志"}],"nameIdentifiers":[{}]}]},"item_26_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Oshima, Makoto","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yamada, Koji","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Endo, Satoshi","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_26_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA1242354X","subitem_source_identifier_type":"NCID"}]},"item_26_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"モンテカルロ碁は知識表現を用いずに棋力を成立させるコンピュータ囲碁である。精度を保つには膨大な計算量を必要とするが、ゲーム木に対して適切な枝刈りを行うことで効率化が可能である。本実験では既存のゲーム知識、特に置石が周囲に与える影響を表したポテンシャルモデルを枝刈りに利用することで計算量の削減を行った。枝刈り方法は設定の異なる4種用意し、其々の削減の効果・傾向を測った。最も効果の高い枝刈り方法では18%、また観測された特性を考慮し、2種の枝刈り方法を組み合わせることで23%まで計算量が削減可能となった。但し限定された環境下での結果の為、実戦の観点に立ち、更に調整を加え試行を重ねる必要がある。","subitem_description_type":"Other"}]},"item_26_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Monte Carlo go is the computer go which satisfy the strength without the knowledge expressions of igo. Monte Carlo go needs an enormous computational complexity to keep the precision. Though, reductions of the computational complexity are possible by proper pruning for the igo game tree. In this study, we tackled the reduction of the computational complexity by the pruning for the igo game tree using the potential model which was the knowledge expression of igo. In this experiment, 4 kind of pruning were tried and measured. The best one pruning reached an 18% reduction of the computational complexity and the proper combination of two pruning reached a 23% reduction of the computational complexity. We only showed a probability of the potential model pruning this time. Thus we need to tackle many trials in different environments after this.","subitem_description_type":"Other"}]},"item_26_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"208","bibliographic_titles":[{"bibliographic_title":"情報科学技術フォーラム講演論文集"}],"bibliographicPageStart":"201","bibliographicIssueDates":{"bibliographicIssueDate":"2011-09-07","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"10"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"updated":"2025-01-20T16:21:00.125375+00:00","links":{}}