{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00212444","sets":["1164:3865:10488:10659"]},"path":["10659"],"owner":"44499","recid":"212444","title":["センサベースの人間行動認識における深層学習アンサンブル手法に関する考察"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-08-26"},"_buckets":{"deposit":"f5fe597f-ffdb-4334-9a0c-53f2ad8c6b75"},"_deposit":{"id":"212444","pid":{"type":"depid","value":"212444","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"センサベースの人間行動認識における深層学習アンサンブル手法に関する考察","author_link":["541974","541975","541976","541977"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"センサベースの人間行動認識における深層学習アンサンブル手法に関する考察"},{"subitem_title":"Consideration of Ensemble Deep Learning Method for Sensor-based Human Activity Recognition","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"センサ","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-08-26","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"福井大学大学院工学研究科"},{"subitem_text_value":"福井大学大学院工学研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Engineering, University of Fukui","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Engineering, University of Fukui","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/212444/files/IPSJ-MBL21100025.pdf","label":"IPSJ-MBL21100025.pdf"},"date":[{"dateType":"Available","dateValue":"2023-08-26"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MBL21100025.pdf","filesize":[{"value":"1.1 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":"35"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"cc917ac7-4d43-488c-977e-835b984749ca","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"長谷川, 達人"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"近藤, 和真"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Tatsuhito, Hasegawa","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kazuma, Kondo","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11851388","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":"2188-8817","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"センサを用いた人間行動認識において深層学習を用いた手法が数多く提案されている．中でも，深層学習とアンサンブル学習を併用する手法は強力な成果を発揮している．一方，アンサンブル学習を行うにはデータの分割や複数モデルを学習するなどの様々な手続きを要し，手間と計算コストがかかる．本研究では，行動認識を対象に深層学習のアンサンブル手法を解析することを通じて，単一モデルを End-to-End で訓練するだけでアンサンブルモデルと同等の推定精度を実現する手法の実現可能性を考察する．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Many methods based on deep learning have been proposed for sensor-based human activity recognition. Methods that combine deep learning and ensemble learning have especially shown powerful results. On the other hand, ensemble learning requires various procedures, such as data partitioning and training multiple models, which are time-consuming and computationally expensive. In this study, we analyze ensemble methods of deep learning for activity recognition, and examine the feasibility of a method that achieves estimation accuracy equivalent to that of ensemble models by simply training a single model in an end-to-end manner.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"9","bibliographic_titles":[{"bibliographic_title":"研究報告モバイルコンピューティングと新社会システム（MBL）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-08-26","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"25","bibliographicVolumeNumber":"2021-MBL-100"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":212444,"updated":"2025-01-19T17:31:07.636472+00:00","links":{},"created":"2025-01-19T01:13:23.374818+00:00"}