Abstract
We present a practical realization of a Radial Basis Function Network for handwritten digits recognition task. Inspired from regularization theory and Parzen windows non parametric estimator, Radial Basis Function networks are tested for a classification task. Reduction of the number of hidden nodes which is an important and necessary step to obtain a computationally tractable network is made using an original technique. A comparison is made with the k-nearest neighbour and Parzen windows methods. Results appear better for the network at a much lower computational cost.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Poggio T., Girosi F., Networks for approximation and learning.Proceedings of the IEEE, Vol 78, No. 9,1990.
Girosi F., Poggio T., Networks and the best approximation property. Biological Cybernetic 63, 169–176, 1990.
Richard M. D., Lippman R. P., Neural Networks Classifiers estimate a posteriori Probabilities. Neural Computation, 4, 461–483, 1991.
Lee Y., Handwritten recognition using K Nearest-Neighbour, Radial Basis Function and Backpropagation Neural Networks. Neural Computation, 3, 440–449, 1991.
Moody J., Darken C. J., Fast learning in Networks of locally tuned processing units, Neural Computation, 1, 281–294, 1989.
Ng, Lipmann R.P., A comparative study of the practical characteristics of neural networks and conventional pattern clasifiers, in Neural Information Processing Systems 3, 1991, D.S. Touretzky, ed. Morgan Kaufmann, San Mateo, Ca.
Geman S, Bienenstock E., Boursat R., Neural Networks and the bias variance dilemma, Neural Computation, 4, 1–58, 1992.
Gluksman H. A., Classification of Mixed Font alphabetics by characteristics loci., 1st annual IEEE Computer Conference, 138–141, 1967.
Gaillat G., Berthod M., Panorama des techniques d'extraction de traits caractéristiques en lecture, optique des caractères, Revue Technique THOMSON-CSF, Vol 11, No 4, 1979.
Specht D. F., Probabilistic Neural Networks, Neural Networks, vol. 3, 109–118, 1990.
Musavi M. T., Ahmed W., Chand K. H., Faris K. B., Hummels D. M., On the Training of Radial Basis Function Classifiers, Neural Networks, Vol 5, 595–605, 1992.
Hausler D., Decision Theoretic Generalization of the PAC Model for Neural Net and Other Learning Applications, Technical Report, UCSC-CRL-91-02, 1991.
White H, Conectionist Non Parametric Regression: Multilayer feedforward Networks can learn Arbitrary Mappings, Neural Networks, Vol. 3, 535–549, 1990.
Devroye L.,Automatic Pattern Recognition: A study of the Probability of Error, IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 10, No.4, 1988.
Chiu S.T., Bandwith Selection for Kernel Density Estimation, The Annals of Statistics, vol. 19, no. 4, 1883–1905, 1991.
Somorjai R.L., Ali M.K. An efficient algorithm for estimating dimensionalities, Can. J. Chem. 66, 979, 1988.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1993 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lemarié, B. (1993). Practical realization of a radial basis function network for handwritten digit recognition. In: Mira, J., Cabestany, J., Prieto, A. (eds) New Trends in Neural Computation. IWANN 1993. Lecture Notes in Computer Science, vol 686. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-56798-4_136
Download citation
DOI: https://doi.org/10.1007/3-540-56798-4_136
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-56798-1
Online ISBN: 978-3-540-47741-9
eBook Packages: Springer Book Archive