Rigorous Results on the Hopfield Model of Neural Networks

Authors

  • Anton Bovier lnstitut fur Angewandte Analysis lind Stochastik
  • Véronique Gayrard Centre de Physique Theorique, CNRS

DOI:

https://doi.org/10.11606/resimeusp.v1i2/3.74568

Keywords:

Disordered systems, Neural networks, Memory capacity, Random graphs.

Abstract

Vile review some recent rigorous results in the theory of neural networks, and in particular on the thermodynamic properties of the Hopfield model. In this context, the model is treated as a Cllrie-Weiss model with random interact, ions and large deviation t,echniques are applied, The tractability of the random interactions depends strongly on how the number, M. of stored patterns scales with the size N. of the system.

Downloads

Downloads

Published

1994-05-01

Issue

Vol. 1 No. 2/3 (1994)

Section

Contents

License

Copyright (c) 1994 Anton Bovier, Véronique Gayrard

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

How to Cite

Rigorous Results on the Hopfield Model of Neural Networks. (1994). Resenhas Do Instituto De Matemática E Estatística Da Universidade De São Paulo, 1(2/3), 161-172. https://doi.org/10.11606/resimeusp.v1i2/3.74568