Elsevier

Physics Letters A

Volume 125, Issue 1, 26 October 1987, Pages 32-34
Physics Letters A

A maximum overlap neural network for pattern recognition

https://doi.org/10.1016/0375-9601(87)90513-5Get rights and content

Abstract

The problem of recognizing noisy patterns is trivially solved by identifying the key pattern which has the largest overlap with the input. We show that a procedure that does this can be implemented in a simple layered neural network. Recognition is free of spurious states, the network never reaches a state of total confusion. Processing time scales as log M, where M is the number of stored patterns.

References (3)

  • J.J. HopfieldJ.J. HopfieldD.J. Amit et al.

    Phys. Rev. A

    (1985)
There are more references available in the full text version of this article.

Cited by (0)

1

Permanent address: Service de Physique Théorique, CEN Saclay, 91191 Gif-sur-Yvette Cedex, France.

View full text