{"updated":"2025-01-19T16:39:47.885288+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00214329","sets":["1164:2836:10501:10733"]},"path":["10733"],"owner":"44499","recid":"214329","title":["秘匿分解データを用いた新しい機械学習"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-12-13"},"_buckets":{"deposit":"fcc5e236-a4a3-4fc1-9b2e-83769258e583"},"_deposit":{"id":"214329","pid":{"type":"depid","value":"214329","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"秘匿分解データを用いた新しい機械学習","author_link":["549964","549963","549962","549961"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"秘匿分解データを用いた新しい機械学習"},{"subitem_title":"New Machine Learning Method with using Secure Divided Data","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"機械学習・予測モデル","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-12-13","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"長崎大学"},{"subitem_text_value":"鹿児島大学"},{"subitem_text_value":"鹿児島大学"},{"subitem_text_value":"中央大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Nagasaki University","subitem_text_language":"en"},{"subitem_text_value":"Kagoshima University","subitem_text_language":"en"},{"subitem_text_value":"Kagoshima University","subitem_text_language":"en"},{"subitem_text_value":"Chuo 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/214329/files/IPSJ-DPS21189011.pdf","label":"IPSJ-DPS21189011.pdf"},"date":[{"dateType":"Available","dateValue":"2023-12-13"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-DPS21189011.pdf","filesize":[{"value":"1.0 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":"34"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"b8448f7d-b71a-4772-963b-efbe07228089","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":"宮島, 洋文"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"重井, 徳貴"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"宮島, 廣美"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"白鳥, 則郎"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10116224","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":"2188-8906","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"持続可能な社会の実現に向けて，SDGs (Sustainable Development Goals) に対する取り組みが世界中で模索されている．日本では Society 5.0 が目指す超スマート社会の構築をテーマの一つとして掲げている．超スマート社会は，サイバー空間とフィジカル空間（現実社会）の高度な融和を目指すものであり，ビッグデータを AI が解析することで，個人のニーズに合った有効な情報がより迅速に現実社会にもたらされる．それでは，ビッグデータに対するプライバシーの侵害や管理強化を防ぐ超スマート社会はどのように構築すれば良いのであろうか．この問題の解決のためには，サイバー空間でのビックデータのプライバシーを保護する AI の解析手法の開発が重要になる．これまでに，この分野では，ユーザにとって安心・安全な AI の解析法としての機械学習の開発の観点から，1）秘密共有＋ SMC (Secure Multiparty Computation)，2）準同型暗号化，3）連合学習，等に関する研究が行われているが，学習法の利活用に対する価値とプライバシーのリスクに対するバランスが高度にとれた方法は知られていない．これらの背景を踏まえて，本論文では，簡易秘匿分解データを用いた分散処理による学習法を提案する．この方法では，あらかじめ個々のデータを乱数を使って複数に分解し，それぞれの断片を各サーバに保存する．学習は断片データを使って，各サーバでの部分計算と中央サーバでの統合計算を繰り返し実行する．提案法の利点としては，学習データをそのまま使うことがないことによるセキュリティの向上と，連合学習と同様に分散処理による機械学習法の実現により多くの問題への利活用が容易となる．この提案法に基づいて，機械学習の応用例として，簡易秘匿分解データを用いた分散処理によるBP (Back Propagation) のアルゴリズムを提案し，その有効性を示す．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告マルチメディア通信と分散処理（DPS）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-12-13","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"11","bibliographicVolumeNumber":"2021-DPS-189"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"created":"2025-01-19T01:15:09.068839+00:00","id":214329,"links":{}}