{"links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00225537","sets":["1164:2822:11181:11182"]},"path":["11182"],"owner":"44499","recid":"225537","title":["OS-ELMを用いたオンライン逐次型グラフ分散表現学習法"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-03-16"},"_buckets":{"deposit":"4aabce9c-f4af-4d19-9188-bc3759cee400"},"_deposit":{"id":"225537","pid":{"type":"depid","value":"225537","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"OS-ELMを用いたオンライン逐次型グラフ分散表現学習法","author_link":["596992","596991","596989","596990"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"OS-ELMを用いたオンライン逐次型グラフ分散表現学習法"},{"subitem_title":"An Online Sequential Graph Distributed Representation Learning Method using OS-ELM","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"深層学習","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2023-03-16","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"慶應義塾大学大学院理工学研究科"},{"subitem_text_value":"慶應義塾大学大学院理工学研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Science and Technology, Keio University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School 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/225537/files/IPSJ-EMB23062030.pdf","label":"IPSJ-EMB23062030.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-EMB23062030.pdf","filesize":[{"value":"2.5 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"42"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"3600e529-266a-4fa2-ab4c-1dcf0087f77a","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 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":"伊藤, 響"}],"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":"Hibiki, Ito","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":"グラフのデータ構造を持つデータを扱う際には隣接行列として読み込んで利用することが一般的であるが，グラフ中の各ノードを分散表現と呼ばれるベクトルに変換して利用することも提案されている．分散表現を学習する過程においてグラフ中のノードの接続状況を考慮したデータが使用されるため，得られる分散表現はそれらが表す各ノードの特徴を表現したものとなっている．そのため，分散表現同士で演算を行ったり，機械学習アルゴリズムに掛けることで，情報を抽出することが可能である．しかし，現状のグラフ分散表現学習法ではバッチ学習を行っているため，データ傾向がドリフトしてしまう実世界のグラフデータの学習には適していないと考えられる．そこで本論文では，OS-ELM というオンライン逐次学習アルゴリズムを用いたグラフ分散表現学習のオンライン学習化を提案する．また，学習を高速化するため，Negative Sampling と呼ばれる高速化手法も組み合わせることも提案する．これらの手法の組み合わせにより，データ傾向が変化した場合においてもモデル破棄の必要が無く，変化を素早く捉えたグラフ分散表現の学習が可能となる．また，分散表現の利用やオンライン学習化によってメモリ使用量も削減される．評価結果から，Negative Sampling 適用後においてもメモリ使用量は 54.4% 削減可能であることがわかった．また，学習は 8.44 倍高速化できた．精度についてもオリジナルと同程度となることがわかった．","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":"2023-03-16","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"30","bibliographicVolumeNumber":"2023-EMB-62"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"created":"2025-01-19T01:25:02.708859+00:00","updated":"2025-01-19T12:46:38.102365+00:00","id":225537}