{"updated":"2025-01-20T06:03:44.691558+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00175954","sets":["581:8417:8429"]},"path":["8429"],"owner":"11","recid":"175954","title":["畳み込みニューラルネットワークを用いた囲碁における1局の棋譜からの棋力推定"],"pubdate":{"attribute_name":"公開日","attribute_value":"2016-11-15"},"_buckets":{"deposit":"fc67c9d6-aaf0-4d05-a5d2-cfae51578cb8"},"_deposit":{"id":"175954","pid":{"type":"depid","value":"175954","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"畳み込みニューラルネットワークを用いた囲碁における1局の棋譜からの棋力推定","author_link":["368755","368758","368757","368754","368756","368759"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"畳み込みニューラルネットワークを用いた囲碁における1局の棋譜からの棋力推定"},{"subitem_title":"Estimating Player's Strength by CNN from One Game Record of Go","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"[特集：ゲームプログラミング] 囲碁，Convolutional Neural Network，棋力推定，棋力分類，レート値","subitem_subject_scheme":"Other"}]},"item_type_id":"2","publish_date":"2016-11-15","item_2_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"電気通信大学／日本学術振興会"},{"subitem_text_value":"電気通信大学"},{"subitem_text_value":"電気通信大学"}]},"item_2_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"The University of Electro Communications / Japan Society for the Promotion of Science","subitem_text_language":"en"},{"subitem_text_value":"The University of Electro Communications","subitem_text_language":"en"},{"subitem_text_value":"The University of Electro Communications","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"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/175954/files/IPSJ-JNL5711005.pdf","label":"IPSJ-JNL5711005.pdf"},"date":[{"dateType":"Available","dateValue":"2018-11-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-JNL5711005.pdf","filesize":[{"value":"2.4 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"8"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"f716a4fe-a4e3-40b9-a8e3-b8ffc1a78313","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2016 by the Information Processing Society of Japan"}]},"item_2_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"荒木, 伸夫"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"保木, 邦仁"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"村松, 正和"}],"nameIdentifiers":[{}]}]},"item_2_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Nobuo, Araki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kunihito, Hoki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Masakazu, Muramatsu","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_2_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00116647","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_2_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7764","subitem_source_identifier_type":"ISSN"}]},"item_2_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"囲碁のプロ棋士は，1局の棋譜を見ればプレイヤの棋力が分かるといわれている．本稿では，畳み込みニューラルネットワーク（Convolutional Neural Network; CNN）を使用し，1局の囲碁の棋譜より，プレイヤの棋力を推定する手法を提案する．プレイヤのレート値を推定する実験と，プレイヤを上級/中級/初級にクラス分けする実験を行った．提案手法を実装して囲碁クエストの13路盤棋譜データを用いて学習させて実験したところ，レート値を推定する手法としては従来手法より平均自乗誤差が小さくなった．また，クラス分類する実験においては，1度CNNを用いてレート値を推定してからその値に応じてクラス分けを行う手法と，最初からクラス分類をCNNに学習させる手法の2種類を提案し，それぞれ長所と短所があることを確かめた．","subitem_description_type":"Other"}]},"item_2_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"It is said that any professional player can estimate a player's strength accurately by looking at just one game record. We propose to use Convolutional Neural Network (CNN) to estimate a Go player's strength from only one game record. We perform two experiments: (i) to estimate a player's rating, and (ii) to classify a player into three classes in strength. We use game records provided by GoQuest to train CNN. For estimating ratings, we compare our method with an existing method to find that our method gives a smaller average mean squared error than that of the existing method. For the classification, we compare two methods: (i) the method that classify a player according to the rating predicted by the CNN, and (ii) the method that trains CNN directly to classify a player based on just one game record. We observed that the two methods have different strong points and weak points.","subitem_description_type":"Other"}]},"item_2_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"2373","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌"}],"bibliographicPageStart":"2365","bibliographicIssueDates":{"bibliographicIssueDate":"2016-11-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"11","bibliographicVolumeNumber":"57"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"created":"2025-01-19T00:45:40.510567+00:00","id":175954,"links":{}}