{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00210539","sets":["1164:2822:10563:10564"]},"path":["10564"],"owner":"44499","recid":"210539","title":["OS-ELMを用いたFPGA向け軽量ファインチューニング手法"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-03-18"},"_buckets":{"deposit":"6331ad33-39f8-4365-924f-2c55a51dd402"},"_deposit":{"id":"210539","pid":{"type":"depid","value":"210539","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"OS-ELMを用いたFPGA向け軽量ファインチューニング手法","author_link":["533300","533299","533302","533298","533297","533301"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"OS-ELMを用いたFPGA向け軽量ファインチューニング手法"},{"subitem_title":"A Fine-Tuning Method using OS-ELM for FPGAs","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"機械学習","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-03-18","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"慶應義塾大学理工学部"},{"subitem_text_value":"慶應義塾大学大学院理工学研究科"},{"subitem_text_value":"慶應義塾大学理工学部"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":" Faculty of Science and Technology, Keio University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Science and Technology, Keio University","subitem_text_language":"en"},{"subitem_text_value":" Faculty of Science and Technology, Keio 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/210539/files/IPSJ-EMB21056041.pdf","label":"IPSJ-EMB21056041.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-EMB21056041.pdf","filesize":[{"value":"1.9 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"42"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"83a24cac-4396-4143-b72f-f2a5fc60b1fc","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 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":"山田, 赳也"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"塚田, 峰登"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"松谷, 宏紀"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Takeya, Yamada","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Mineto, Tsukada","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroki, Matsutani","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA12149313","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":"2188-868X","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"訓練データと実環境データの特徴量の性質が異なっていることがあり，あらゆる現場環境下を想定し，訓練データを全て用意することは現実的に難しい場合がある．訓練データと実環境データにおける正常パターンの特徴に差異がある場合，訓練済みモデルを，実環境のデータを用いて再度訓練するファインチューニング手法が利用される．しかし，学習モデルの再訓練や訓練データの収集は必ずしも容易ではない．そこで，本論文では，畳み込み Auto Encoder の Encode 部分とオンライン逐次学習アルゴリズム OS-ELM (Online  Sequential  Extreme  Learning  Machine) を連結することで，現場で軽量にファインチューニングする手法を提案する．実験では，手書き文字認識データセット MNIST を用いて，畳み込み Auto Encoder を事前学習しておき，現場のデータとしてそれらを左右にシフトさせたデータを用いた．評価では，ファインチューニングしない通常のニューラルネットワーク，畳み込み Auto Encoder，畳み込み Auto  Encoder と OS-ELM を組みわせたモデル，提案手法の精度を比較した．その結果，提案手法を用いて現場でファインチューニングした提案手法の精度は訓練データで学習した畳み込み Auto Encoder よりも 4.1% 高くなった．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告組込みシステム（EMB）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-03-18","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"41","bibliographicVolumeNumber":"2021-EMB-56"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":210539,"updated":"2025-01-19T18:07:25.541815+00:00","links":{},"created":"2025-01-19T01:11:45.302376+00:00"}