Abstract
The aim of the paper is to combine the generalization power of classical density estimation techniques with the efficiency of VLSI-friendly, constructive algorithms. This is important for VLSI-implementations of classification networks in the case of noisy data to avoid the well-known “overfitting” effects. The method consists of three steps 1) estimation of the probability densities for each class, 2) discretization of the input space and 3) describing the resulting classification regions using only easy implementable boolean AND and OR gates and comparisons. The “noisy spirals” classification problem, a noisy variant of the “two spirals” benchmark, is used for demonstration.
on leave from Institut für Theoretische Physik I, Universität Münster. J.L. was supported by the British-German ARC-programme
on leave of absence from the “Politehnica” University of Bucharest. V.B. is a HC&M Research Fellow of the EC and is financed by the Commission of the European Communities under contract ERBCHBICT941741.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Beiu, V., Peperstraete, J.A., Vandewalle, J. and Lauwereins, R. (1993). Efficient Decomposition of Comparison and Its Applications. In Verleysen, M.(ed.): ESANN’93, D facto, Brussels, Belgium, 1993, 45–50.
Beiu, V., Peperstraete, J.A., Vandewalle J., and Lauwereins, R. (1994) Learning from Examples and VLSI Implementation of Neural Networks. In R. Trappl (ed.): Cybernetics and System Research 94, World Scientific Publishing, Singapore, 1994, 1767–1774.
Beiu, V. and Taylor, J.G. (1995). Optimal VLSI Implementations of Neural Networks: VLSI-Friendly Learning Algorithms. The Applied Decision Technologies ADT’95, London, UK, 3–5 April, 1995.
Fiesler, E. (1994) Comparative Bibliography of Ontogenic Neural Networks. ICANN94, Proceedings, Springer, London, 793–796.
Lang K.J., and Witbrock M.J. (1988). Learning to tell two spirals apart. Proc. Connectionists Models Summer School, Morgan Kauffmann, San Mateo, 52–59.
Poggio, T., and Girosi, F.A. (1990). Networks for Approximation and Learning. Proceedings of the IEEE 78, September 1990, 1481–1497.
Silverman, B.W. (1986) Density Estimation. Chapman & Hall, London.
Smieja, F.J. (1993). Neural Network Constructive Algorithm: Trading Generalisation for Learning Efficiency? Circuits, Systems, and Signal Processing, 12(2), 331–374.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1995 Springer-Verlag London Limited
About this paper
Cite this paper
Lemm, J.C., Beiu, V., Taylor, J.G. (1995). Density Estimation as a Preprocessing Step for Constructive Algorithms. In: Kappen, B., Gielen, S. (eds) Neural Networks: Artificial Intelligence and Industrial Applications. Springer, London. https://doi.org/10.1007/978-1-4471-3087-1_41
Download citation
DOI: https://doi.org/10.1007/978-1-4471-3087-1_41
Publisher Name: Springer, London
Print ISBN: 978-3-540-19992-2
Online ISBN: 978-1-4471-3087-1
eBook Packages: Springer Book Archive