{"created":"2025-01-19T01:28:04.044665+00:00","updated":"2025-01-19T11:39:58.522568+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00228936","sets":["1164:2592:11085:11360"]},"path":["11360"],"owner":"44499","recid":"228936","title":["時相論理仕様を満足するマルチエージェントシステムの深層強化学習"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-11-09"},"_buckets":{"deposit":"ec087258-bd3e-4394-8386-20341ae373ad"},"_deposit":{"id":"228936","pid":{"type":"depid","value":"228936","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"時相論理仕様を満足するマルチエージェントシステムの深層強化学習","author_link":["614829","614825","614827","614828","614824","614826"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"時相論理仕様を満足するマルチエージェントシステムの深層強化学習"},{"subitem_title":"Deep Reinforcement Learning for Multi-Agent Systems with Temporal Logic Specifications","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2023-11-09","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"北海道大学"},{"subitem_text_value":"北海道大学"},{"subitem_text_value":"北海道大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Hokkaido University","subitem_text_language":"en"},{"subitem_text_value":"Hokkaido University","subitem_text_language":"en"},{"subitem_text_value":"Hokkaido University","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/228936/files/IPSJ-AL23195016.pdf","label":"IPSJ-AL23195016.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-AL23195016.pdf","filesize":[{"value":"1.1 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"9"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"776509e4-8fd2-4b1c-9c9f-aa85dc94c510","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"寺嶋, 啓太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"小林, 孝一"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"山下, 裕"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Keita, Terashima","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Koichi, Kobayashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yuh, Yamashita","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN1009593X","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":"2188-8566","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"マルチエージェントシステムでは，共通目標の達成のためにエージェント群が如何に協調した行動をとるかが課題となる．著者らは以前に，時相論理仕様を満たす協調型マルチエージェント強化学習法を提案した．アグリゲータを導入した分散的な学習手法によって，エージェント数の増加に伴う状態空間爆発および学習効率の低下という問題点を解消した．本論文では，この手法に深層強化学習を適用する．その際，表形式の評価関数を用いた強化学習法では実装が困難な連続状態空間の環境を対象とした学習を行う．その後，連続空間で表された監視問題を例として，深層強化学習におけるアグリゲータの効果を検証する．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In multi-agent systems, the challenge is how a group of agents collaborate to achieve a common goal. In our previous work, we propose cooperative multi-agent reinforcement learning methods with temporal logic specifications. The distributed method with an aggregator solves the technical issues of state-space explosion and low learning efficiency as the number of agents grows. In this paper, we apply deep reinforcement learning to our methods. We consider learning methods for continuous state-space environments, which are difficult to implement in reinforcement learning methods using a table-style evaluation function. Then, we verify the effectiveness of an aggregator in deep reinforcement learning using an example of a surveillance problem represented in a continuous space.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"5","bibliographic_titles":[{"bibliographic_title":"研究報告アルゴリズム（AL）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2023-11-09","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"16","bibliographicVolumeNumber":"2023-AL-195"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":228936,"links":{}}