{"id":241070,"updated":"2025-01-19T07:45:11.817627+00:00","links":{},"created":"2025-01-19T01:45:37.514488+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00241070","sets":["1164:3206:11566:11793"]},"path":["11793"],"owner":"44499","recid":"241070","title":["画像分類ネットワークにおける学習データセットの出力層での分布の可視化と解析"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-11-22"},"_buckets":{"deposit":"78535e33-4119-455a-815f-40c095f56000"},"_deposit":{"id":"241070","pid":{"type":"depid","value":"241070","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"画像分類ネットワークにおける学習データセットの出力層での分布の可視化と解析","author_link":["663047","663045","663044","663046"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"画像分類ネットワークにおける学習データセットの出力層での分布の可視化と解析"}]},"item_type_id":"4","publish_date":"2024-11-22","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"埼玉工業大学"},{"subitem_text_value":"福岡大学"},{"subitem_text_value":"青森大学"},{"subitem_text_value":"東京電機大学"}]},"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/241070/files/IPSJ-CG24196011.pdf","label":"IPSJ-CG24196011.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CG24196011.pdf","filesize":[{"value":"2.3 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"28"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"7e3de862-2191-4bc0-b7bf-aabc1364d6ac","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 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":"望月, 義彦"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"藤木, 淳"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"江口, 脩"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"大山, 航"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10100541","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":"2188-8949","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"画像分類ネットワークの学習では，用いられるデータセットのデータ件数が十分な認識性能を得るためには不足していると考えられる場合に，データセットの画像に変形や加工を行うことによって画像のバリエーションを増やす「データ拡張」が行われる．しかし，データ拡張にはドメイン知識を考慮する必要があったり，データセット内の全てのデータに画像処理を行うため，データ拡張に大きな計算コストが必要となるという問題がある．データセットの特徴空間における分布がクラス間の識別率にどのように影響するかを定量的，もしくは定性的に評価できれば，この評価指標を用いた効率的なデータ拡張や，より認識性能の高いネットワークの再設計への活用が期待できる．そこで本研究では，データセットが出力層でどのように分布するかを可視化し，識別境界付近のデータ拡張の分布や誤識別の様子などを調べる．また，出力層から入力データを復元する逆写像学習により，識別境界の決定に重要なデータがどのようであるものかの分析を試みる．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告コンピュータグラフィックスとビジュアル情報学（CG）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2024-11-22","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"11","bibliographicVolumeNumber":"2024-CG-196"}]},"relation_version_is_last":true,"weko_creator_id":"44499"}}