Abstract
This paper proposes a new adaptive learning algorithm of network structure aimed at improving RBF network generalization ability. The algorithm determines the initial number and center vectors of network hidden units by using forward selective clustering algorithm with decaying radius, and then adjusts them by using cluster sample transform algorithm based on impurity and variance and gets the final center vectors. The determination of widths of hidden units considers both the dispersivity of inner samples and the distance between clusters. Thus we get the final hidden structure. After determining the hidden structure, the back-propagation algorithm is used to train the weights between the hidden layer and output layer. The experiment of two spirals problem proves that our algorithm has higher generalization ability indeed.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Whitehead B A. Cooperative-competitive genetic evolution of radial basis function centers and widths fortime series prediction[J].IEEE Trans on Neural Networks,1996, 7(4):869–880.
Bianchini, M., Frasconi, P., and Gori, M. (1995). Learning without local minima in radial basis function networks. IEEE Transactions on Neural Networks,1995,6(3):749–756.
M. J. Orr, Regularization in the selection of RBF centres, Neural Computation, vol. 7, no. 3, pp. 606–623, 1995.
V. Vapnik, the Nature of Statistical Learning Theory, Springer Verlag, 1995.
Tom Carter. An introduction to information theory and entropy. http://cogs.csustan.edu/~tom/information, June 2000
J. E. Moody and J. Utans. Architecture selection strategies for neural networks: Application to corporate bond rating prediction. In A. N. Refenes, editor, Neural Networks in the Capital Markets. John Wiley & Sons, 1994.
J.C. Platt. A resource allocating network for function interpolation. Neural Computation, 3:213–225, 1991.
Fritzke, B. (1994b). Growing cells structures-a self-organizing network for unsupervised and supervised learning. Neural Networks, 7(9):1441–1460.1993.
Y.Y. Yao, S.K.M. Wong, and C.J. Butz. On information-theoretic measures of attribute importance. In N. Zhong and L. Zhou, editors, Proceedings of the Third Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD’99), pages 133–137, Beijing, China, April 1999.
K. Lang and M. Witbrock, Learning to tell two spirals apart, in Proc.
Lu YW, Sundararajan N, Saratchandran P. Performance evaluation of sequential mininal radial basis function (RBF) neural network learning algorithm[J].IEEE Trans on Neural Networks,1998,9(6):308–317.
Michael R. Berthold and Jay Diamond. Boosting the Performance of RBF Networks with Dynamic Decay Adjustment. Advances in Neural information processings 7, 1995.
Ales Leonardis and Horst Bischof. An efficient MDL-Based construction of RBF networks. Neural Networks, 11(5):963–973, July 1998.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sun, J., Shen, RM., Yang, F. (2002). An Adaptive Learning Algorithm Aimed at Improving RBF Network Generalization Ability. In: McKay, B., Slaney, J. (eds) AI 2002: Advances in Artificial Intelligence. AI 2002. Lecture Notes in Computer Science(), vol 2557. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-36187-1_32
Download citation
DOI: https://doi.org/10.1007/3-540-36187-1_32
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-00197-3
Online ISBN: 978-3-540-36187-9
eBook Packages: Springer Book Archive