Convergence index for BPN training

The Back Propagation Network (BPN) is one of the most widely used neural networks. It has a distinct training phase and then it is put to use. It is observed that certain training sets would Get trained and certain others would not get trained in a BPN. This is an inherent property of the training s...

Full description

Bibliographic Details
Main Author: Gunasekaran, S
Format: Article
Published: 2003
Subjects:
Online Access:http://shdl.mmu.edu.my/2536/
_version_ 1848790082204467200
author Gunasekaran, S
author_facet Gunasekaran, S
author_sort Gunasekaran, S
building MMU Institutional Repository
collection Online Access
description The Back Propagation Network (BPN) is one of the most widely used neural networks. It has a distinct training phase and then it is put to use. It is observed that certain training sets would Get trained and certain others would not get trained in a BPN. This is an inherent property of the training set and the BPN architecture. This paper proposes a new training index that may be evaluated after a certain number of training epochs and would indicate the ability of the BPN to train for the existing topology and training set. The index is evaluated for several logic gates and the results are presented. (C) 2003 Elsevier B.V. All rights reserved.
first_indexed 2025-11-14T18:06:57Z
format Article
id mmu-2536
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:06:57Z
publishDate 2003
recordtype eprints
repository_type Digital Repository
spelling mmu-25362011-08-22T06:05:10Z http://shdl.mmu.edu.my/2536/ Convergence index for BPN training Gunasekaran, S QA75.5-76.95 Electronic computers. Computer science The Back Propagation Network (BPN) is one of the most widely used neural networks. It has a distinct training phase and then it is put to use. It is observed that certain training sets would Get trained and certain others would not get trained in a BPN. This is an inherent property of the training set and the BPN architecture. This paper proposes a new training index that may be evaluated after a certain number of training epochs and would indicate the ability of the BPN to train for the existing topology and training set. The index is evaluated for several logic gates and the results are presented. (C) 2003 Elsevier B.V. All rights reserved. 2003-10 Article NonPeerReviewed Gunasekaran, S (2003) Convergence index for BPN training. Neurocomputing, 55 (3-4). pp. 711-719. ISSN 09252312 http://dx.doi.org/10.1016/S0925-2312(03)00368-0 doi:10.1016/S0925-2312(03)00368-0 doi:10.1016/S0925-2312(03)00368-0
spellingShingle QA75.5-76.95 Electronic computers. Computer science
Gunasekaran, S
Convergence index for BPN training
title Convergence index for BPN training
title_full Convergence index for BPN training
title_fullStr Convergence index for BPN training
title_full_unstemmed Convergence index for BPN training
title_short Convergence index for BPN training
title_sort convergence index for bpn training
topic QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/2536/
http://shdl.mmu.edu.my/2536/
http://shdl.mmu.edu.my/2536/