Bounds on the covariance matrix of the Sherrington-Kirkpatrick model

TL;DR

This paper proves that the expected operator norm of the covariance matrix in the SK model with no external field and temperature below 1 is bounded by a constant depending only on temperature, answering an open question.

Contribution

It establishes a new bound on the covariance matrix's operator norm, improving upon previous logarithmic bounds by deriving an approximate formula via TAP equations.

Findings

Expected operator norm is bounded by a temperature-dependent constant.

Diverging lower bounds are shown at critical and low temperatures.

Provides an approximate covariance matrix formula through differentiation of TAP equations.

Abstract

We consider the Sherrington-Kirkpatrick model with no external field and inverse temperature $\beta<1$ and prove that the expected operator norm of the covariance matrix of the Gibbs measure is bounded by a constant depending only on $\beta$. This answers an open question raised by Talagrand, who proved a bound of $C(\beta) (\log n)^8$. Our result follows by establishing an approximate formula for the covariance matrix which we obtain by differentiating the TAP equations and then optimally controlling the associated error terms. We complement this result by showing diverging lower bounds on the operator norm, both at the critical and low temperatures.