Abstract
We review two new algorithms for learning in neural networks of Boolean units. The first applies to the problem of associative memory: Hopfield model or perception. The algorithm optimizes the stability of learned patterns, which enlarges the size of the basins of attraction. The second algorithm builds a multilayer feedforward network: it allows one to learn an arbitrary mapping input → output. The convergence of the growth process is guaranteed. The generalization properties look very promising.
This paper gives a short review of some work done at ENS on learning in neural networks, basically in collaboration with W. Krauth and J.P. Nadal. I shall not give any details (the reader is referred to the original papers), but I shall rather point out the basic ideas and some results.
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.
References
Minsky, M., Papert, S.: Perceptions, MIT Press, Cambridge, MA, (1969).
Krauth, W., Mezard, M.: Learning algorithms with optimal stability in neural networks. J. Phys. A, 20, L 745, (1987).
Forrest, B.M.: J. Phys. A21, 245 (1988).
Krauth, W., Nadal, J.P., Mezard, M.: Basins of attraction in a perceptron-like neural network. Complex Systems 2, 387 (1988).
Krauth, W., Nadal, J.P., Mezard, M.: The roles of stability and symmetry in the dynamics of neural networks. J. Phys. A, 21, 2995–3011, (1988).
Gardner, E.: Europhys. Lett. 4, 481 (1987)
Gardner, E. Derrida, B.: Optimal Storage Properties of Neural Network Models. In J. Phys. A, 21, 271–284,(1988).
Gardner, E., Stroud, N., Wallace, D.: Training with Noise, and the Storage of Correlated Patterns in a Neural Network Model, Edinburgh Preprint 87/394, University of Edinburgh, (1987).
Pceppel G. and Krey U.: Europhys. Lett. 4, 481, (1987).
Mézard, M., Nadal, J.P.: Learning in feed-forward layered networks: the tiling algorithm. J. Phys. A, 22, 2191–2203, (1989).
Rujan P., Marchand, M.: Learning by activating neurons: a new approach to learning in neural networks. KFA Jülich preprint, to appear in Complex Systems, (1988).
Carnevali, P., Patarnello, S.: Europhys. Lett. 4, 1199, (1987).
Denker, J., Schwartz, D., Wittner, B., Solla, S., Hopfield, J.J. Howard, R., Jackel, L.: Automatic Learning, Rule Extraction and Generalization. In Complex Systems vol 1, 877–922, (1987).
Nadal, J.P.: Study of a growth algorithm for a feedforward network. In Intl Journal of Neural Networks, 1, 55–59, (1989).
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1990 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mézard, M. (1990). Learning algorithms in neural networks: recent results. In: Soulié, F.F., Hérault, J. (eds) Neurocomputing. NATO ASI Series, vol 68. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-76153-9_8
Download citation
DOI: https://doi.org/10.1007/978-3-642-76153-9_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-76155-3
Online ISBN: 978-3-642-76153-9
eBook Packages: Springer Book Archive