Please use this identifier to cite or link to this item: https://doi.org/10.1109/TNN.2004.830801
DC FieldValue
dc.titleNew dynamical optimal learning for linear multilayer FNN
dc.contributor.authorTan, K.C.
dc.contributor.authorTang, H.J.
dc.date.accessioned2014-06-17T02:58:42Z
dc.date.available2014-06-17T02:58:42Z
dc.date.issued2004-11
dc.identifier.citationTan, K.C., Tang, H.J. (2004-11). New dynamical optimal learning for linear multilayer FNN. IEEE Transactions on Neural Networks 15 (6) : 1562-1568. ScholarBank@NUS Repository. https://doi.org/10.1109/TNN.2004.830801
dc.identifier.issn10459227
dc.identifier.urihttp://scholarbank.nus.edu.sg/handle/10635/56795
dc.description.abstractThis letter presents a new dynamical optimal learning (DOL) algorithm for three-layer linear neural networks and investigates its generalization ability. The optimal learning rates can be fully determined during the training process. The mean squared error (mse) is guaranteed to be stably decreased and the learning is less sensitive to initial parameter settings. The simulation results illustrate that the proposed DOL algorithm gives better generalization performance and faster convergence as compared to standard error back propagation algorithm. © 2004 IEEE.
dc.description.urihttp://libproxy1.nus.edu.sg/login?url=http://dx.doi.org/10.1109/TNN.2004.830801
dc.sourceScopus
dc.subjectBack propagation
dc.subjectDynamical optimal learning (DOL)
dc.subjectFeedforward neural networks (FNN)
dc.subjectStability
dc.typeArticle
dc.contributor.departmentELECTRICAL & COMPUTER ENGINEERING
dc.description.doi10.1109/TNN.2004.830801
dc.description.sourcetitleIEEE Transactions on Neural Networks
dc.description.volume15
dc.description.issue6
dc.description.page1562-1568
dc.description.codenITNNE
dc.identifier.isiut000224929600019
Appears in Collections:Staff Publications

Show simple item record
Files in This Item:
There are no files associated with this item.

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.