File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
  • Find via Find It@HKUL
Supplementary

Article: B-spline recurrent neural network and its nonlinear modelling

TitleB-spline recurrent neural network and its nonlinear modelling
基本樣條循環神經網絡及其非線性建模
Authors
KeywordsRecurrent neural network (循環神經網絡)
B-spline function (基本樣條函數)
Adaptive learning algorithm (自適應學習算法)
Optimal learning rate (最優學習步長)
Issue Date1999
PublisherNortheastern University (東北大學). The Journal's web site is located at http://kzyjc.periodicals.net.cn/
Citation
Control and Decision, 1999, v. 14 n. 5, p. 469-472 How to Cite?
控制與決策, 1999, v. 14 n. 5, p. 469-472 How to Cite?
AbstractA new recurrent neural network based on B-spline function approximation is presented. The network can be easily trained and its training converges more quickly. Moreover, in order to overcome the slowness of training caused by constant learning rate, an adaptive weight updating algorithm used for training the recurrent network is proposed and an optimal estimate of learning rate is given. Examples are given to show that the optimal learning rate can be used to train the B-spline recurrent neural network, and this recurrent neural network can be applied to modelling of a nonlinear dynamic system. 提出一種新的基于基本樣條逼近的循環神經網絡,該網絡易于訓練且收斂速度快。此外為克服定長學習步長訓練速度慢的問題,提出一種用于該網絡訓練的自適應權值更新算法,給出了學習步長的最優估計。該最優學習步長的選擇可用于基本樣條循環神經網絡的訓練以及對非線性系統的建模。
Persistent Identifierhttp://hdl.handle.net/10722/75987
ISSN
2023 SCImago Journal Rankings: 0.288

 

DC FieldValueLanguage
dc.contributor.authorJin, H-
dc.contributor.authorChan, CW-
dc.contributor.authorZhang, HY-
dc.date.accessioned2010-09-06T07:16:30Z-
dc.date.available2010-09-06T07:16:30Z-
dc.date.issued1999-
dc.identifier.citationControl and Decision, 1999, v. 14 n. 5, p. 469-472-
dc.identifier.citation控制與決策, 1999, v. 14 n. 5, p. 469-472-
dc.identifier.issn1001-0920-
dc.identifier.urihttp://hdl.handle.net/10722/75987-
dc.description.abstractA new recurrent neural network based on B-spline function approximation is presented. The network can be easily trained and its training converges more quickly. Moreover, in order to overcome the slowness of training caused by constant learning rate, an adaptive weight updating algorithm used for training the recurrent network is proposed and an optimal estimate of learning rate is given. Examples are given to show that the optimal learning rate can be used to train the B-spline recurrent neural network, and this recurrent neural network can be applied to modelling of a nonlinear dynamic system. 提出一種新的基于基本樣條逼近的循環神經網絡,該網絡易于訓練且收斂速度快。此外為克服定長學習步長訓練速度慢的問題,提出一種用于該網絡訓練的自適應權值更新算法,給出了學習步長的最優估計。該最優學習步長的選擇可用于基本樣條循環神經網絡的訓練以及對非線性系統的建模。-
dc.languagechi-
dc.publisherNortheastern University (東北大學). The Journal's web site is located at http://kzyjc.periodicals.net.cn/-
dc.relation.ispartofControl and Decision-
dc.relation.ispartof控制與決策-
dc.subjectRecurrent neural network (循環神經網絡)-
dc.subjectB-spline function (基本樣條函數)-
dc.subjectAdaptive learning algorithm (自適應學習算法)-
dc.subjectOptimal learning rate (最優學習步長)-
dc.titleB-spline recurrent neural network and its nonlinear modelling-
dc.title基本樣條循環神經網絡及其非線性建模-
dc.typeArticle-
dc.identifier.openurlhttp://library.hku.hk:4550/resserv?sid=HKU:IR&issn=1001-0920&volume=14 &issue=5&spage=469&epage=472&date=1999&atitle=B-spline+recurrent+neural+network+and+its+nonlinear+modellingen_HK
dc.identifier.emailChan, CW: mechan@hkucc.hku.hk-
dc.identifier.authorityChan, CW=rp00088-
dc.identifier.hkuros49558-
dc.identifier.volume14-
dc.identifier.issue5-
dc.identifier.spage469-
dc.identifier.epage472-
dc.publisher.placeChina (中國)-
dc.identifier.issnl1001-0920-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats