{"links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00235730","sets":["6504:11678:11697"]},"path":["11697"],"owner":"44499","recid":"235730","title":["対数線形化された無限混合正規分布に基づく未知クラス推定確率ニューラルネットの構造最適化"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-03-01"},"_buckets":{"deposit":"07bee6ae-44aa-40e3-a268-8bfdb65d4bc2"},"_deposit":{"id":"235730","pid":{"type":"depid","value":"235730","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"対数線形化された無限混合正規分布に基づく未知クラス推定確率ニューラルネットの構造最適化","author_link":["644187","644190","644188","644189"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"対数線形化された無限混合正規分布に基づく未知クラス推定確率ニューラルネットの構造最適化"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ソフトウェア科学・工学","subitem_subject_scheme":"Other"}]},"item_type_id":"22","publish_date":"2024-03-01","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_22_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"横浜国大"},{"subitem_text_value":"神奈川県立産業技術総合研究所"},{"subitem_text_value":"横浜国大"},{"subitem_text_value":"横浜国大"}]},"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/235730/files/IPSJ-Z86-7K-06.pdf","label":"IPSJ-Z86-7K-06.pdf"},"date":[{"dateType":"Available","dateValue":"2024-07-03"}],"format":"application/pdf","filename":"IPSJ-Z86-7K-06.pdf","filesize":[{"value":"1.4 MB"}],"mimetype":"application/pdf","accessrole":"open_date","version_id":"6e263b61-aead-48fe-976b-6e4e15475c01","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_22_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"南部, 穣汰"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"迎田, 隆幸"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"柏木, 僚太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"島, 圭介"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_22_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00349328","subitem_source_identifier_type":"NCID"}]},"item_22_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"我々の研究グループでは，学習時に想定しない未知事象を含めた多クラス分類を実現できるオープンセット認識(OSR)手法(NACGMN)を提案しており，複数の確率密度関数を内包する特殊な構造により，極少量の学習データからも高精度な分類が可能となる．一方，NACGMNにはハイパーパラメータの経験的設定が必要で，教師なし学習問題に適用できないという課題がある．本研究ではNACGMNの事前学習に無限混合正規分布（IGMM）を導入し，ハイパーパラメータの学習的獲得，教師ラベルの自動生成によって上記の課題を解消する．また，IGMMに含まれる冗長な確率演算の対数線形化による，高速化も試みる．実験では筋電位信号等の分類タスクに提案法を応用し，有効性を検証した．","subitem_description_type":"Other"}]},"item_22_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"242","bibliographic_titles":[{"bibliographic_title":"第86回全国大会講演論文集"}],"bibliographicPageStart":"241","bibliographicIssueDates":{"bibliographicIssueDate":"2024-03-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"2024"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"updated":"2025-01-19T09:32:33.725403+00:00","created":"2025-01-19T01:37:22.564058+00:00","id":235730}