{"created":"2021-03-01T06:17:20.393070+00:00","id":791,"links":{},"metadata":{"_buckets":{"deposit":"ada71253-9f90-4204-8224-21122bcf1604"},"_deposit":{"id":"791","owners":[],"pid":{"revision_id":0,"type":"depid","value":"791"},"status":"published"},"_oai":{"id":"oai:repository.dl.itc.u-tokyo.ac.jp:00000791","sets":["30:31","9:10:14"]},"item_2_alternative_title_1":{"attribute_name":"その他のタイトル","attribute_value_mlt":[{"subitem_alternative_title":"Systems Solving Inverse Problems by Learning Nonlinear Gain"}]},"item_2_biblio_info_7":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"1994-07","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"7","bibliographicPageEnd":"1354","bibliographicPageStart":"1345","bibliographicVolumeNumber":"77","bibliographic_titles":[{"bibliographic_title":"電子情報通信学会論文誌. D-II, 情報・システム, II-情報処理"}]}]},"item_2_description_5":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"対象となるシステムの目標出力を実現する入力の値を求める問題は逆問題と呼ばれており，その解法システムを神経回路等の学習要素によって獲得する研究が行われている．直接逆モデリングや順逆モデリング等によって獲得された逆モデルを利用する手法が提案されているが，それぞれ問題がある．筆者らは，対象システムの線形化モデルの逆モデルを直接逆モデリングによって学習し，それを利用して出力誤差をフィードバックし，ニュ一トン法に似た反復改良により解を計算する出力フィードバック型逆 モデルを提案した．解の初期値のランダム探索と組み合わせることにより，高精度の逆問題の解を計算できる． しかしながら，出力フィードバック型逆 モデルへの入力は，解の推定値の現在値と出力誤差の両方であ り，その入力次元は大きく，学習に時間がかかる． また，反復改良計算の時間も長い．本論文では，非線形ゲインを学習によって獲得し，出力誤差に非線形ゲインを乗じることによって，出力誤差のフィードバック回路を形成する解法システムを提案し，その性能と学習の効果をロボットアームの逆運動学問題を 解くシミュレーションによって確認する．","subitem_description_type":"Abstract"}]},"item_2_full_name_3":{"attribute_name":"著者別名","attribute_value_mlt":[{"nameIdentifiers":[{"nameIdentifier":"3037","nameIdentifierScheme":"WEKO"}],"names":[{"name":"Oyama, Eimei"}]},{"nameIdentifiers":[{"nameIdentifier":"3038","nameIdentifierScheme":"WEKO"}],"names":[{"name":"Tachi, Susumu"}]}]},"item_2_publisher_20":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"電子情報通信学会"}]},"item_2_relation_25":{"attribute_name":"関係URI","attribute_value_mlt":[{"subitem_relation_type_id":{"subitem_relation_type_id_text":"http://www.ieice.org/eng/","subitem_relation_type_select":"URI"}}]},"item_2_rights_12":{"attribute_name":"権利","attribute_value_mlt":[{"subitem_rights":"copyright©1994 IEICE, 許諾番号07RB0130"}]},"item_2_source_id_10":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN1007132X","subitem_source_identifier_type":"NCID"}]},"item_2_source_id_8":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"09151923","subitem_source_identifier_type":"ISSN"}]},"item_2_subject_15":{"attribute_name":"日本十進分類法","attribute_value_mlt":[{"subitem_subject":"007","subitem_subject_scheme":"NDC"}]},"item_2_text_4":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"機械技術研究所"},{"subitem_text_value":"東京大学先端科学技術研究センター"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"大山, 英明"}],"nameIdentifiers":[{"nameIdentifier":"3035","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"舘, 暲"}],"nameIdentifiers":[{"nameIdentifier":"3036","nameIdentifierScheme":"WEKO"}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2017-05-30"}],"displaytype":"detail","filename":"tachi_1994_033d.pdf","filesize":[{"value":"655.7 kB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"tachi_1994_033d.pdf","url":"https://repository.dl.itc.u-tokyo.ac.jp/record/791/files/tachi_1994_033d.pdf"},"version_id":"405b04f7-ab74-41b6-9076-f49d77219a7e"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"神経回路","subitem_subject_scheme":"Other"},{"subitem_subject":"逆問題","subitem_subject_scheme":"Other"},{"subitem_subject":"出力フィードバツク型逆モデル","subitem_subject_scheme":"Other"},{"subitem_subject":"非線形ゲインの学習","subitem_subject_scheme":"Other"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"journal article","resourceuri":"http://purl.org/coar/resource_type/c_6501"}]},"item_title":"非線形ゲインを学習する逆問題解法システム","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"非線形ゲインを学習する逆問題解法システム"}]},"item_type_id":"2","owner":"1","path":["31","14"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-01-06"},"publish_date":"2017-01-06","publish_status":"0","recid":"791","relation_version_is_last":true,"title":["非線形ゲインを学習する逆問題解法システム"],"weko_creator_id":"1","weko_shared_id":null},"updated":"2022-12-19T03:41:54.050899+00:00"}