{"updated":"2025-01-19T08:02:49.134582+00:00","links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00240215","sets":["6164:6165:6640:11802"]},"path":["11802"],"owner":"44499","recid":"240215","title":["深層アンサンブルモデルのMulti-Narrow化が分類精度に与える影響"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-06-19"},"_buckets":{"deposit":"2bf2ce2c-f777-4914-843a-00589512d3f5"},"_deposit":{"id":"240215","pid":{"type":"depid","value":"240215","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"深層アンサンブルモデルのMulti-Narrow化が分類精度に与える影響","author_link":["658704","658705"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"深層アンサンブルモデルのMulti-Narrow化が分類精度に与える影響"},{"subitem_title":"The impact of Multi-Narrowing of Deep Ensemble Models on classification accuracy.","subitem_title_language":"en"}]},"item_type_id":"18","publish_date":"2024-06-19","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_18_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"福井大学大学院"},{"subitem_text_value":"福井大学大学院"}]},"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/240215/files/IPSJ-DICOMO2024099.pdf","label":"IPSJ-DICOMO2024099.pdf"},"date":[{"dateType":"Available","dateValue":"2026-06-19"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-DICOMO2024099.pdf","filesize":[{"value":"1.5 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":"44"}],"accessrole":"open_date","version_id":"37c12555-52fb-47d5-83f2-713a54c2bc37","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_18_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"田中, 大聖"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"長谷川, 達人"}],"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_18_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本研究では，Convolutional Neural Network（CNN）に対し，チャネルの分割・並列化を施すことでアンサンブルモデルを構築する手法「Multi-Narrow化」が示す精度変化について検証と考察を行う．本手法は，行動認識タスクや画像分類タスクの一部条件下においてモデルの性能を向上させることが明らかになっている．しかし，画像分類タスクにおいては分割数が4より多い場合については検証が行われておらず，性能が向上する原理についても未解明である．本研究では，Multi-Narrow化の度合いとデータセットの規模を変化させたときの，精度向上量の変化について検証した．実験の結果，データセットの規模が小さい場合，Multi-Narrow化したモデルはしていないものと比べて高い精度を示すことが明らかになった．また，データセットの規模によって，Multi-Narrow化による精度上昇量は大きく変化することを示し，その要因について考察を行った．","subitem_description_type":"Other"}]},"item_18_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"738","bibliographic_titles":[{"bibliographic_title":"マルチメディア，分散，協調とモバイルシンポジウム2024論文集"}],"bibliographicPageStart":"731","bibliographicIssueDates":{"bibliographicIssueDate":"2024-06-19","bibliographicIssueDateType":"Issued"},"bibliographicVolumeNumber":"2024"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":240215,"created":"2025-01-19T01:44:17.006780+00:00"}