Annals of Applied Probability

Extreme Value Behavior in the Hopfield Model

Anton Bovier and David M. Mason

Full-text: Open access

Abstract

We study a Hopfield model whose number of patterns M grows to infinity with the system size N,in such a way that M(N)2 log M(N)/N tends to zero. In this model the unbiased Gibbs state in volume N can essentially be decomposed into M(N) pairs of disjoint measures. We investigate the distributions of the corresponding weights,and show,in particular, that these weights concentrate for any given N very closely to one of the pairs, with probability tending to 1. Our analysis is based upon a new result on the asymptotic distribution of order statistics of certain correlated exchangeable random variables.

Article information

Source
Ann. Appl. Probab., Volume 11, Number 1 (2001), 91-120.

Dates
First available in Project Euclid: 27 August 2001

Permanent link to this document
https://projecteuclid.org/euclid.aoap/998926988

Digital Object Identifier
doi:10.1214/aoap/998926988

Mathematical Reviews number (MathSciNet)
MR1825461

Zentralblatt MATH identifier
1024.82015

Subjects
Primary: 82B44: Disordered systems (random Ising models, random Schrödinger operators, etc.) 60G70: Extreme value theory; extremal processes 60K356

Keywords
Hopfield model extreme values order statistics metastates chaotic size dependence

Citation

Bovier, Anton; Mason, David M. Extreme Value Behavior in the Hopfield Model. Ann. Appl. Probab. 11 (2001), no. 1, 91--120. doi:10.1214/aoap/998926988. https://projecteuclid.org/euclid.aoap/998926988


Export citation