Decay of covariance for gradient models with non-convex potential

TL;DR

This paper analyzes the decay of covariances in non-convex gradient models on lattices, showing they closely resemble Gaussian free field covariances with faster decay corrections, using advanced renormalization techniques.

Contribution

It extends the renormalization group method to observables in non-convex gradient models, providing detailed covariance decay analysis.

Findings

Covariances match Gaussian free field up to faster decay terms.

Established decay rates for covariances in non-convex gradient models.

Extended renormalization group techniques to observables.

Abstract

We consider gradient models on the lattice $Z^d$. These models serve as effective models for interfaces and are also known as continuous Ising models. The height of the interface is modelled by a random field with an energy which is a non-convex perturbation of the quadratic interaction. We are interested in the Gibbs measure with tilted boundary condition $u$ at inverse temperature $\beta$ of this model. In this paper we present a fine analysis of the covariance of the gradient field. We show that the covariances of the Gibbs distribution agree with the covariance of the Gaussian free field up to terms which decay at a faster algebraic rate. The key tool is the extension of the renormalisation group method to observables as developed in [BBS15a].