Abstract
We propose herein a neural network based on curved kernels constituing an anisotropic family of functions and a learning rule to automatically tune the number of needed kernels to the frequency of the data in the input space. The model has been tested on two case studies of approximation problems known to be difficult and gave good results in comparision with traditional radial basis function (RBF) netwoks. Those examples illustrate the fact that curved kernels can locally adapt themselves to match with the observation space regularity.
This is a preview of subscription content, log in via an institution to check access.
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
Darken, C., Moody, J.: Fast learning in networks of locally-tuned processing units. Neural Computation, 1(2):281–294, 1989.
Poggio, T., Girosi, F.: Networks for approximation and learning. Proc. of the IEEE, Vol 78, No 9, September, pp. 1481–1497, 1990.
Benam, M., Tomasini, L.: Competitive and self-organizing algorithms based on the minimization of an information criterion. International Conference on Artificial Neural Networks, 1991.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bourret, P., Pelletier, B. (2001). Curved Kernel Neural Network for Functions Approximation. In: Mira, J., Prieto, A. (eds) Bio-Inspired Applications of Connectionism. IWANN 2001. Lecture Notes in Computer Science, vol 2085. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45723-2_9
Download citation
DOI: https://doi.org/10.1007/3-540-45723-2_9
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42237-2
Online ISBN: 978-3-540-45723-7
eBook Packages: Springer Book Archive