File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Acceleration of Levenberg-Marquardt Training of Neural Networks with Variable Decay Rate

TitleAcceleration of Levenberg-Marquardt Training of Neural Networks with Variable Decay Rate
Authors
KeywordsComputers
Computer networks
Issue Date2003
PublisherIEEE.
Citation
Proceedings Of The International Joint Conference On Neural Networks, 2003, v. 3, p. 1873-1878 How to Cite?
AbstractIn the application of the standard Levenherg-Marquardt training process of neural networks, error oscillations are frequently observed and they usually aggravate on approaching the required accuracy. In this paper, a modified Levenberg-Marquardt method based on variable decay rate in each iteration is proposed in order to reduce such error oscillations. Through a certain variation of the decay rate, the time required for training of neural networks is cut down to less than half of that required in the standard Levenberg-Marquardt method. Several numerical examples are given to show the effectiveness of the proposed method.
Persistent Identifierhttp://hdl.handle.net/10722/47064
ISSN
References

 

DC FieldValueLanguage
dc.contributor.authorChen, TCen_HK
dc.contributor.authorHan, DJen_HK
dc.contributor.authorAu, FTKen_HK
dc.contributor.authorTham, LGen_HK
dc.date.accessioned2007-10-30T07:05:54Z-
dc.date.available2007-10-30T07:05:54Z-
dc.date.issued2003en_HK
dc.identifier.citationProceedings Of The International Joint Conference On Neural Networks, 2003, v. 3, p. 1873-1878en_HK
dc.identifier.issn1098-7576en_HK
dc.identifier.urihttp://hdl.handle.net/10722/47064-
dc.description.abstractIn the application of the standard Levenherg-Marquardt training process of neural networks, error oscillations are frequently observed and they usually aggravate on approaching the required accuracy. In this paper, a modified Levenberg-Marquardt method based on variable decay rate in each iteration is proposed in order to reduce such error oscillations. Through a certain variation of the decay rate, the time required for training of neural networks is cut down to less than half of that required in the standard Levenberg-Marquardt method. Several numerical examples are given to show the effectiveness of the proposed method.en_HK
dc.format.extent305430 bytes-
dc.format.extent1760 bytes-
dc.format.extent2058 bytes-
dc.format.mimetypeapplication/pdf-
dc.format.mimetypetext/plain-
dc.format.mimetypetext/plain-
dc.languageengen_HK
dc.publisherIEEE.en_HK
dc.relation.ispartofProceedings of the International Joint Conference on Neural Networksen_HK
dc.rights©2003 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.en_HK
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License-
dc.subjectComputersen_HK
dc.subjectComputer networksen_HK
dc.titleAcceleration of Levenberg-Marquardt Training of Neural Networks with Variable Decay Rateen_HK
dc.typeConference_Paperen_HK
dc.identifier.openurlhttp://library.hku.hk:4550/resserv?sid=HKU:IR&issn=1098-7576&volume=3&spage=1873&epage=1878&date=2003&atitle=Acceleration+of+Levenberg-Marquardt+training+of+neural+networks+with+variable+decay+rateen_HK
dc.identifier.emailAu, FTK:francis.au@hku.hken_HK
dc.identifier.emailTham, LG:hrectlg@hkucc.hku.hken_HK
dc.identifier.authorityAu, FTK=rp00083en_HK
dc.identifier.authorityTham, LG=rp00176en_HK
dc.description.naturepublished_or_final_versionen_HK
dc.identifier.doi10.1109/IJCNN.2003.1223693en_HK
dc.identifier.scopuseid_2-s2.0-0141794666en_HK
dc.identifier.hkuros93063-
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-0141794666&selection=ref&src=s&origin=recordpageen_HK
dc.identifier.volume3en_HK
dc.identifier.spage1873en_HK
dc.identifier.epage1878en_HK
dc.identifier.scopusauthoridChen, TC=16634751900en_HK
dc.identifier.scopusauthoridHan, DJ=7403219487en_HK
dc.identifier.scopusauthoridAu, FTK=7005204072en_HK
dc.identifier.scopusauthoridTham, LG=7006213628en_HK

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats