Abstract
We present a method for analyzing the behavior of RBFs in an on-line scenario which provides a description of the learning dynamics without invoking the thermodynamic limit. Our analysis is based on a master equation that describes the dynamics of the weight space probability density for any value of the input space dimension. Because the transition probability appearing in the master equation cannot be written in closed form, some approximate form of the dynamics is developed. We assume a arbitrary small learning rate (small noise) and we derive in this limit the dynamic evolution of the means and the variances of the net weights. The analytic results are then confirmed by simulations.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
T. H. Heskes and B. Kappen, Mathematical Foundations of Neural Networks Taylor Ed., Amsterdam 1993. Elsevier, pages 199–233. Physical Review A 44, 1991
Leen T K, Schottky B and Saad D 1998, Advanced in Neural Information Systems Vol. 10, ed. by Jordan, Kearns, Solla (Cambridge, MA: MIT Press) p 301
D. Bedeaux, K. Lakatos-Lindberg and K. Shuler, Journal of mathematical Physics 12, 2116–2123, 1971.
J. Freeman and D. Saad 1997, Physical Review E 56, 907–918.
Saad D (ed) 1998 Online learning in neural networks London: Cambridge University Press. Saad D and Solla S A 1995 Phys. Rev. Lett. 74 4337.
Scarpetta S 1998, Ph. D. thesis, Salerno University
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag London Limited
About this paper
Cite this paper
Marinaro, M., Scarpetta, S. (1999). Dynamics of On-Line Learning in Radial Basis Function Neural Networks. In: Marinaro, M., Tagliaferri, R. (eds) Neural Nets WIRN Vietri-99. Perspectives in Neural Computing. Springer, London. https://doi.org/10.1007/978-1-4471-0877-1_11
Download citation
DOI: https://doi.org/10.1007/978-1-4471-0877-1_11
Publisher Name: Springer, London
Print ISBN: 978-1-4471-1226-6
Online ISBN: 978-1-4471-0877-1
eBook Packages: Springer Book Archive