English

The Hopfield Model with Superlinearly Many Patterns

Probability 2012-10-15 v3

Abstract

We study the Hopfield model where the ratio α\alpha of patterns to sites grows large. We prove that the free energy with inverse temperature β\beta and external field BB behaves like βα+γ\beta\sqrt\alpha+\gamma, where γ\gamma is the limiting free energy of the Sherrington-Kirkpatrick model with inverse temperature 2β\sqrt2\beta and external field BB.

Keywords

Cite

@article{arxiv.1108.4771,
  title  = {The Hopfield Model with Superlinearly Many Patterns},
  author = {James Y. Zhao},
  journal= {arXiv preprint arXiv:1108.4771},
  year   = {2012}
}

Comments

9 pages; Statistics and Probability Letters (2012)

R2 v1 2026-06-21T18:54:30.644Z