Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure

This work presents two novel approaches to determine optimum growing multi-experts network (GMN) structure. The first method called direct method deals with expertise domain and levels in connection with local experts. The growing neural gas (GNG) algorithm is used to cluster the local experts. The...

Full description

Saved in:
Bibliographic Details
Main Authors: Loo, C.K., Rajeswari, M., Rao, M.V.C.
Format: Article
Published: 2004
Subjects:
Online Access:http://eprints.um.edu.my/5179/
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1353276
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.um.eprints.5179
record_format eprints
spelling my.um.eprints.51792013-03-21T01:33:57Z http://eprints.um.edu.my/5179/ Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure Loo, C.K. Rajeswari, M. Rao, M.V.C. T Technology (General) This work presents two novel approaches to determine optimum growing multi-experts network (GMN) structure. The first method called direct method deals with expertise domain and levels in connection with local experts. The growing neural gas (GNG) algorithm is used to cluster the local experts. The concept of error distribution is used to apportion error among the local experts. After reaching the specified size of the network, redundant experts removal algorithm is invoked to prune the size of the network based on the ranking of the experts. However, GMN is not ergonomic due to too many network control parameters. Therefore, a self-regulating GMN (SGMN) algorithm is proposed. SGMN adopts self-adaptive learning rates for gradient-descent learning rules. In addition, SGMN adopts a more rigorous clustering method called fully self-organized simplified adaptive resonance theory in a modified form. Experimental results show SGMN obtains comparative or even better performance than GMN in four benchmark examples, with reduced sensitivity to learning parameters setting. Moreover, both GMN and SGMN outperform the other neural networks and statistical models. The efficacy of SGMN is further justified in three industrial applications and a control problem. It provides consistent results besides holding out a profound potential and promise for building a novel type of nonlinear model consisting of several local linear models. 2004 Article PeerReviewed Loo, C.K. and Rajeswari, M. and Rao, M.V.C. (2004) Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure. Neural Networks, IEEE Transactions on Neural Networks, 15 (6). pp. 1378-1395. ISSN 1045-9227 http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1353276
institution Universiti Malaya
building UM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaya
content_source UM Research Repository
url_provider http://eprints.um.edu.my/
topic T Technology (General)
spellingShingle T Technology (General)
Loo, C.K.
Rajeswari, M.
Rao, M.V.C.
Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
description This work presents two novel approaches to determine optimum growing multi-experts network (GMN) structure. The first method called direct method deals with expertise domain and levels in connection with local experts. The growing neural gas (GNG) algorithm is used to cluster the local experts. The concept of error distribution is used to apportion error among the local experts. After reaching the specified size of the network, redundant experts removal algorithm is invoked to prune the size of the network based on the ranking of the experts. However, GMN is not ergonomic due to too many network control parameters. Therefore, a self-regulating GMN (SGMN) algorithm is proposed. SGMN adopts self-adaptive learning rates for gradient-descent learning rules. In addition, SGMN adopts a more rigorous clustering method called fully self-organized simplified adaptive resonance theory in a modified form. Experimental results show SGMN obtains comparative or even better performance than GMN in four benchmark examples, with reduced sensitivity to learning parameters setting. Moreover, both GMN and SGMN outperform the other neural networks and statistical models. The efficacy of SGMN is further justified in three industrial applications and a control problem. It provides consistent results besides holding out a profound potential and promise for building a novel type of nonlinear model consisting of several local linear models.
format Article
author Loo, C.K.
Rajeswari, M.
Rao, M.V.C.
author_facet Loo, C.K.
Rajeswari, M.
Rao, M.V.C.
author_sort Loo, C.K.
title Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
title_short Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
title_full Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
title_fullStr Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
title_full_unstemmed Novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
title_sort novel direct and self-regulating approaches to determine optimum growing multi-experts network structure
publishDate 2004
url http://eprints.um.edu.my/5179/
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1353276
_version_ 1643687506630148096
score 13.160551