Dynamic tunneling based regularization in feedforward neural networks

This paper presents a new regularization method based on dynamic tunneling for enhancing generalization capability of multilayered neural networks. The proposed method enables escape through undesired sub-optimal solutions on the composite error surface by means of dynamic tunneling. Undesired sub-o...

Full description

Bibliographic Details
Main Authors: Singh, Y.P., RoyChowdhury, Pinaki
Format: Article
Language:English
Published: 2001
Subjects:
Online Access:http://shdl.mmu.edu.my/2685/
http://shdl.mmu.edu.my/2685/1/1926.pdf
_version_ 1848790123139825664
author Singh, Y.P.
RoyChowdhury, Pinaki
author_facet Singh, Y.P.
RoyChowdhury, Pinaki
author_sort Singh, Y.P.
building MMU Institutional Repository
collection Online Access
description This paper presents a new regularization method based on dynamic tunneling for enhancing generalization capability of multilayered neural networks. The proposed method enables escape through undesired sub-optimal solutions on the composite error surface by means of dynamic tunneling. Undesired sub-optimal solutions may be increased or introduced from regularized objective function. Hence, the proposed method is capable of enhancing the regularization property without getting stuck at sub-optimal values in search space. The regularization property and escape from the sub-optimal values have been demonstrated through computer simulations on two examples. (C) 2001 Elsevier Science B.V. All rights reserved.
first_indexed 2025-11-14T18:07:36Z
format Article
id mmu-2685
institution Multimedia University
institution_category Local University
language English
last_indexed 2025-11-14T18:07:36Z
publishDate 2001
recordtype eprints
repository_type Digital Repository
spelling mmu-26852014-02-06T04:22:26Z http://shdl.mmu.edu.my/2685/ Dynamic tunneling based regularization in feedforward neural networks Singh, Y.P. RoyChowdhury, Pinaki QA75.5-76.95 Electronic computers. Computer science This paper presents a new regularization method based on dynamic tunneling for enhancing generalization capability of multilayered neural networks. The proposed method enables escape through undesired sub-optimal solutions on the composite error surface by means of dynamic tunneling. Undesired sub-optimal solutions may be increased or introduced from regularized objective function. Hence, the proposed method is capable of enhancing the regularization property without getting stuck at sub-optimal values in search space. The regularization property and escape from the sub-optimal values have been demonstrated through computer simulations on two examples. (C) 2001 Elsevier Science B.V. All rights reserved. 2001-09 Article NonPeerReviewed text en http://shdl.mmu.edu.my/2685/1/1926.pdf Singh, Y.P. and RoyChowdhury, Pinaki (2001) Dynamic tunneling based regularization in feedforward neural networks. Artificial Intelligence, 131 (1-2). pp. 55-71. ISSN 00043702 http://dx.doi.org/10.1016/S0004-3702(01)00112-6 doi:10.1016/S0004-3702(01)00112-6 doi:10.1016/S0004-3702(01)00112-6
spellingShingle QA75.5-76.95 Electronic computers. Computer science
Singh, Y.P.
RoyChowdhury, Pinaki
Dynamic tunneling based regularization in feedforward neural networks
title Dynamic tunneling based regularization in feedforward neural networks
title_full Dynamic tunneling based regularization in feedforward neural networks
title_fullStr Dynamic tunneling based regularization in feedforward neural networks
title_full_unstemmed Dynamic tunneling based regularization in feedforward neural networks
title_short Dynamic tunneling based regularization in feedforward neural networks
title_sort dynamic tunneling based regularization in feedforward neural networks
topic QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/2685/
http://shdl.mmu.edu.my/2685/
http://shdl.mmu.edu.my/2685/
http://shdl.mmu.edu.my/2685/1/1926.pdf