Glauber dynamics in a zero magnetic field and eigenvalue spacing statistics

TL;DR

This paper investigates the eigenvalue spacing statistics of Glauber dynamics across various statistical mechanics models, revealing different regimes from non-universal to G.O.E. statistics and eigenvalue condensation phenomena.

Contribution

It provides a comprehensive analysis of eigenvalue spacing in Glauber matrices for multiple models, highlighting temperature-dependent statistical regimes and eigenvalue condensation effects.

Findings

Eigenvalue statistics are non-universal in 1D Ising model.

High temperature regimes show intermediate between Poisson and G.O.E. statistics.

Eigenvalues condense around integers at low temperatures.

Abstract

We discuss the eigenvalue spacing statistics of the Glauber matrix for various models of statistical mechanics (a one dimensional Ising model, a two dimensional Ising model, a one dimensional model with a disordered ground state, and a SK model with and without a ferromagnetic bias). The dynamics of the one dimensional Ising model are integrable, and the eigenvalue spacing statistics are non-universal. In the other cases, the eigenvalue statistics in the high temperature regime are intermediate between Poisson and G.O.E (with $P(0)$ of the order of $0.5$). In the intermediate temperature regime, the statistics are G.O.E.. In the low temperature regime, the statistics have a peak at $s=0$. In the low temperature regime, and for disordered systems, the eigenvalues condense around integers, due to the fact that the local field on any spin never vanishes. This property is still valid for the Ising model on the Cayley tree, even if it is not disordered. We also study the spacing between the two largest eigenvalues as a function of temperature. This quantity seems to be sensitive to the existence of a broken symmetry phase.