{"created":"2025-01-19T01:29:16.359053+00:00","updated":"2025-01-19T11:23:20.507010+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00229858","sets":["6504:11436:11440"]},"path":["11440"],"owner":"44499","recid":"229858","title":["アテンション機構に基づく複数CNNモデルの統合によるマルチソース転移学習"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-02-16"},"_buckets":{"deposit":"ab579289-e213-4678-b56b-cbb244853b4c"},"_deposit":{"id":"229858","pid":{"type":"depid","value":"229858","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"アテンション機構に基づく複数CNNモデルの統合によるマルチソース転移学習","author_link":["618330","618329"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"アテンション機構に基づく複数CNNモデルの統合によるマルチソース転移学習"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"人工知能と認知科学","subitem_subject_scheme":"Other"}]},"item_type_id":"22","publish_date":"2023-02-16","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_22_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東理大"},{"subitem_text_value":"東理大"}]},"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/229858/files/IPSJ-Z85-2P-03.pdf","label":"IPSJ-Z85-2P-03.pdf"},"date":[{"dateType":"Available","dateValue":"2023-11-17"}],"format":"application/pdf","filename":"IPSJ-Z85-2P-03.pdf","filesize":[{"value":"413.5 kB"}],"mimetype":"application/pdf","accessrole":"open_date","version_id":"5f31f942-80bf-4747-a469-b5ba6d3ff5e1","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 by the Information Processing Society of Japan"}]},"item_22_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"森山, 総太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"中村, 和晃"}],"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_22_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00349328","subitem_source_identifier_type":"NCID"}]},"item_22_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"転移学習において「負の転移」を防ぐ手段の一つとして，異なるソースタスクで学習した複数のモデルを統合し利用するマルチソース転移学習がある．この際の統合は，一般にブースティング等の手法によりdecision-levelで行われるが，本研究では，主に画像認識を対象としたfeature-levelの統合法を提案し，認識精度の向上を目指す．提案手法では，アテンション機構の導入によりモデルごとの注目領域に多様性を持たせ，モデル同士の連携を高める．その際，アテンションマップの値域や方向（空間方向かチャンネル方向か）が統合後モデルの性能に与える影響を実験的に調査し，手法の更なる改善を図る．","subitem_description_type":"Other"}]},"item_22_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"116","bibliographic_titles":[{"bibliographic_title":"第85回全国大会講演論文集"}],"bibliographicPageStart":"115","bibliographicIssueDates":{"bibliographicIssueDate":"2023-02-16","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"2023"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":229858,"links":{}}