{"@context":{"@vocab":"https://cir.nii.ac.jp/schema/1.0/","rdfs":"http://www.w3.org/2000/01/rdf-schema#","dc":"http://purl.org/dc/elements/1.1/","dcterms":"http://purl.org/dc/terms/","foaf":"http://xmlns.com/foaf/0.1/","prism":"http://prismstandard.org/namespaces/basic/2.0/","cinii":"http://ci.nii.ac.jp/ns/1.0/","datacite":"https://schema.datacite.org/meta/kernel-4/","ndl":"http://ndl.go.jp/dcndl/terms/","jpcoar":"https://github.com/JPCOAR/schema/blob/master/2.0/"},"@id":"https://cir.nii.ac.jp/crid/1390866591962855040.json","@type":"Article","productIdentifier":[{"identifier":{"@type":"DOI","@value":"10.9746/sicetr.61.104"}},{"identifier":{"@type":"NDL_BIB_ID","@value":"034055493"}},{"identifier":{"@type":"URI","@value":"http://id.ndl.go.jp/bib/034055493"}},{"identifier":{"@type":"URI","@value":"https://ndlsearch.ndl.go.jp/books/R000000004-I034055493"}},{"identifier":{"@type":"URI","@value":"https://www.jstage.jst.go.jp/article/sicetr/61/3/61_104/_pdf"}}],"resourceType":"学術雑誌論文(journal article)","dc:title":[{"@language":"en","@value":"Construction and Input-to-output Characteristics Evaluation of Compressed Model of Recurrent Neural Networks for Their Stability Analysis"},{"@language":"ja","@value":"再帰型ニューラルネットワークの安定性判別のための圧縮モデルの構築と入出力特性の評価"},{"@language":"ja-Kana","@value":"サイキガタ ニューラルネットワーク ノ アンテイセイ ハンベツ ノ タメ ノ アッシュク モデル ノ コウチク ト ニュウシュツリョク トクセイ ノ ヒョウカ"}],"dc:language":"ja","description":[{"type":"abstract","notation":[{"@language":"en","@value":"<p>This paper proposes a model compression method of reducing the number of nonlinear activation functions of continuous-time recurrent neural networks (RNNs). Ensuring the internal stability of the compressed RNN guarantees that of the original RNN. An error bound between the outputs of the compressed RNN and the original one is derived. Moreover, an optimization problem for reducing the bound is formulated, and it is relaxed to a semi-definite programming problem. Furthermore, it is shown that the proposed model compression method produces a compressed RNN whose output is close to that of the original one as a general tendency. The proposed method is demonstrated on a simple numerical example.</p>"}],"abstractLicenseFlag":"disallow"}],"creator":[{"@id":"https://cir.nii.ac.jp/crid/1410866591962855040","@type":"Researcher","foaf:name":[{"@language":"ja","@value":"湯野 剛史"},{"@language":"en","@value":"YUNO Tsuyoshi"}],"jpcoar:affiliationName":[{"@language":"en","@value":"Faculty/Graduate School of Information Science and Electrical Engineering, Kyushu University"},{"@language":"ja","@value":"九州大学大学院システム情報科学研究院"}]},{"@id":"https://cir.nii.ac.jp/crid/1410866591962855042","@type":"Researcher","foaf:name":[{"@language":"ja","@value":"福地 和真"},{"@language":"en","@value":"FUKUCHI Kazuma"}],"jpcoar:affiliationName":[{"@language":"en","@value":"Faculty/Graduate School of Information Science and Electrical Engineering, Kyushu University"},{"@language":"ja","@value":"九州大学大学院システム情報科学研究院"}]},{"@id":"https://cir.nii.ac.jp/crid/1420845751143878784","@type":"Researcher","personIdentifier":[{"@type":"KAKEN_RESEARCHERS","@value":"80346080"},{"@type":"NRID","@value":"1000080346080"},{"@type":"CINII_AUTHOR_ID","@value":"DA17473528"},{"@type":"URI","@value":"https://ci.nii.ac.jp/author/DA17473528#entity"},{"@type":"URI","@value":"https://viaf.org/viaf/NII%7CDA17473528"},{"@type":"NRID","@value":"9000243899892"},{"@type":"NRID","@value":"9000404645265"},{"@type":"NRID","@value":"9000024286181"},{"@type":"NRID","@value":"9000411825087"},{"@type":"NRID","@value":"9000024261899"},{"@type":"NRID","@value":"9000024103996"},{"@type":"NRID","@value":"9000023116838"},{"@type":"NRID","@value":"9000279889185"},{"@type":"NRID","@value":"9000403960195"},{"@type":"NRID","@value":"9000391813706"},{"@type":"NRID","@value":"9000024873950"},{"@type":"NRID","@value":"9000403960186"},{"@type":"NRID","@value":"9000258683699"},{"@type":"NRID","@value":"9000279889191"},{"@type":"NRID","@value":"9000001058535"},{"@type":"NRID","@value":"9000391988958"},{"@type":"NRID","@value":"9000254772277"},{"@type":"NRID","@value":"9000023034473"},{"@type":"NRID","@value":"9000415418892"},{"@type":"RESEARCHMAP","@value":"https://researchmap.jp/read0192962"},{"@type":"RESEARCHMAP","@value":"https://researchmap.jp/yebihara"}],"foaf:name":[{"@language":"ja","@value":"蛯原 義雄"},{"@language":"en","@value":"EBIHARA Yoshio"}],"jpcoar:affiliationName":[{"@language":"en","@value":"Faculty/Graduate School of Information Science and Electrical Engineering, Kyushu University"},{"@language":"ja","@value":"九州大学大学院システム情報科学研究院"}]}],"publication":{"publicationIdentifier":[{"@type":"PISSN","@value":"04534654"},{"@type":"LISSN","@value":"04534654"},{"@type":"EISSN","@value":"18838189"},{"@type":"NDL_BIB_ID","@value":"000000006577"},{"@type":"ISSN","@value":"04534654"},{"@type":"NCID","@value":"AN00072392"}],"prism:publicationName":[{"@language":"en","@value":"Transactions of the Society of Instrument and Control Engineers"},{"@language":"ja","@value":"計測自動制御学会論文集"},{"@language":"en","@value":"Transaction of SICE"},{"@language":"en","@value":"Transactions of the Society of Instrument and Control Engineers"},{"@language":"en","@value":"TSICE"},{"@language":"ja","@value":"計測自動制御学会論文集"},{"@language":"ja","@value":"計測論文"},{"@language":"ja","@value":"ＳＩＣＥ論文誌"},{"@language":"ja","@value":"ＳＩＣＥ論文集"},{"@language":"ja","@value":"計制論"},{"@language":"ja","@value":"制御論文"},{"@language":"en","@value":"T. SICE"}],"dc:publisher":[{"@language":"en","@value":"The Society of Instrument and Control Engineers"},{"@language":"ja","@value":"公益社団法人 計測自動制御学会"}],"prism:publicationDate":"2025","prism:volume":"61","prism:number":"3","prism:startingPage":"104","prism:endingPage":"114"},"reviewed":"false","url":[{"@id":"http://id.ndl.go.jp/bib/034055493"},{"@id":"https://ndlsearch.ndl.go.jp/books/R000000004-I034055493"},{"@id":"https://www.jstage.jst.go.jp/article/sicetr/61/3/61_104/_pdf"}],"availableAt":"2025","foaf:topic":[{"@id":"https://cir.nii.ac.jp/all?q=recurrent%20neural%20networks","dc:title":"recurrent neural networks"},{"@id":"https://cir.nii.ac.jp/all?q=model%20compression","dc:title":"model compression"},{"@id":"https://cir.nii.ac.jp/all?q=stability%20analysis","dc:title":"stability analysis"},{"@id":"https://cir.nii.ac.jp/all?q=nonlinear%20systems","dc:title":"nonlinear systems"}],"project":[{"@id":"https://cir.nii.ac.jp/crid/1040581301854597120","@type":"Project","projectIdentifier":[{"@type":"KAKEN","@value":"23K20949"},{"@type":"JGN","@value":"JP23K20949"},{"@type":"URI","@value":"https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-23K20949/"}],"notation":[{"@language":"ja","@value":"錐計画に基づく再帰型ニューラルネットワークの安定性解析と最適設計"},{"@language":"en","@value":"Stability Analysis and Optimal Synthesis of Recurrent Neural Networks by Conic Programming"}]}],"relatedProduct":[{"@id":"https://cir.nii.ac.jp/crid/1360011143576663424","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"System analysis via integral quadratic constraints"}]},{"@id":"https://cir.nii.ac.jp/crid/1360022497406592000","@type":"Article","resourceType":"学術雑誌論文(journal article)","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"A Lyapunov-Based Method of Reducing Activation Functions of Recurrent Neural Networks for Stability Analysis"}]},{"@id":"https://cir.nii.ac.jp/crid/1360022497407135232","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"Local Lipschitz Constant Computation of ReLU-FNNs: Upper Bound Computation with Exactness Verification"}]},{"@id":"https://cir.nii.ac.jp/crid/1360294646281860480","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"A Comprehensive Review of Stability Analysis of Continuous-Time Recurrent Neural Networks"}]},{"@id":"https://cir.nii.ac.jp/crid/1360303972394474880","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"Closed-form continuous-time neural networks"}]},{"@id":"https://cir.nii.ac.jp/crid/1360303976305338112","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"On Static O’Shea-Zames-Falb Multipliers for Idempotent Nonlinearities"}]},{"@id":"https://cir.nii.ac.jp/crid/1360585451495252992","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"Stability Analysis of Recurrent Neural Networks by IQC with Copositive Mutipliers"}]},{"@id":"https://cir.nii.ac.jp/crid/1361699995077379072","@type":"Article","relationType":["references"],"jpcoar:relatedTitle":[{"@value":"Convex Optimization"}]},{"@id":"https://cir.nii.ac.jp/crid/1390295802047924992","@type":"Article","resourceType":"学術雑誌論文(journal article)","relationType":["references"],"jpcoar:relatedTitle":[{"@language":"en","@value":"Stability Analysis of Continuous-Time Recurrent Neural Networks by IQC with Copositive Multipliers"},{"@language":"ja","@value":"積分二次制約と共正値マルチプライアを用いた連続時間再帰型ニューラルネットワークの安定性解析"}]}],"dataSourceIdentifier":[{"@type":"JALC","@value":"oai:japanlinkcenter.org:2013944615"},{"@type":"NDL_SEARCH","@value":"oai:ndlsearch.ndl.go.jp:R000000004-I034055493"},{"@type":"CROSSREF","@value":"10.9746/sicetr.61.104"},{"@type":"KAKEN","@value":"PRODUCT-25590387"}]}