Abstract
This paper presents a modification to the incremental learning algorithm originally proposed by Bernd Fritzke.This algorithm is a single stage approach to build a RBF neural network in the training phase. It combines the unsupervised and supervised learning stages to incrementally generate the RBF architecture. The algorithm uses accumulated error information to insert new neurons dynamically in the hidden space. The algorithm maintains constant parameters to control the growth of the network. This may result in non-optimal architectures. Modifications that provide adaptive capabilities to the learning parameters have been proposed in this work. Two benchmark data sets have been used for training. Results show an improved performance over the original incremental learning algorithm.
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
Broomhead, Lowe: Multivariable Function Interpolation and Adaptive Networks. Complex Systems 2, 312–355 (1988)
Moody, T.J., Darken, C.J.: Fast Learning in Networks of Locally Tuned Processing Units. Neural Computation 1, 151–160 (1989)
Poggio, T., Girosi, F.: Networks for Approximation and Learning. Proceedings of the IEEE 78(9), 1481–1497 (1990)
Platt, J.C.: Learning by Combining Memorization and Gradient Descent. J. Neural Information System-3 (1991)
Fahlman, S.E., Lebiere, C.: The Cascade-Correlation Learning Architecture. Advances in Neural Information Processing Systems 2, 524–532 (1990)
Platt, J.C.: A Resource-Allocating Network for Function Interpolation. Neural Computation 3(2), 213–225 (1991)
Fritzke, B.: Supervised Learning with Growing Cell Structures. Advances in Neural Information Processing Systems 6, 255–262 (1994)
Nagabhushan, T.N., Padma, S.K.: Incremental Neural Learning Algorithms for Classification of Spiral Structures. In: Proceedings of the National Conference on Recent Trends in Information Technology, Karpagam Arts & Science College, Coimbatore, pp. 19–23 (2002)
Fritzke, B.: Fast Learning with Incremental RBF Networks. Neural Processing Letters 1(1), 2–5 (1994)
Nagabhushan, T.N., Padma, S.K.: Extended Tower Algorithm for Multi Category Classification, ICICS-PCM 2003, Nanyang Technological University, Singapore (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nagabhushan, T.N., Padma, S.K. (2004). Adaptive Learning in Incremental Learning RBF Networks. In: Pal, N.R., Kasabov, N., Mudi, R.K., Pal, S., Parui, S.K. (eds) Neural Information Processing. ICONIP 2004. Lecture Notes in Computer Science, vol 3316. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30499-9_72
Download citation
DOI: https://doi.org/10.1007/978-3-540-30499-9_72
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23931-4
Online ISBN: 978-3-540-30499-9
eBook Packages: Springer Book Archive