Open Access
Translator Disclaimer
February 2001 Extreme Value Behavior in the Hopfield Model
Anton Bovier, David M. Mason
Ann. Appl. Probab. 11(1): 91-120 (February 2001). DOI: 10.1214/aoap/998926988

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.

Citation

Download Citation

Anton Bovier. David M. Mason. "Extreme Value Behavior in the Hopfield Model." Ann. Appl. Probab. 11 (1) 91 - 120, February 2001. https://doi.org/10.1214/aoap/998926988

Information

Published: February 2001
First available in Project Euclid: 27 August 2001

zbMATH: 1024.82015
MathSciNet: MR1825461
Digital Object Identifier: 10.1214/aoap/998926988

Subjects:
Primary: 60G70 , 60K356 , 82B44

Keywords: chaotic size dependence , Extreme values , Hopfield model , metastates , order statistics

Rights: Copyright © 2001 Institute of Mathematical Statistics

JOURNAL ARTICLE
30 PAGES


SHARE
Vol.11 • No. 1 • February 2001
Back to Top