Distribution of levels in high-dimensional random landscapes

TL;DR

This paper establishes that in high-dimensional discrete structures, the distribution of various random field levels converges to a Gaussian process with specific covariance properties, revealing a universal behavior across models.

Contribution

It proves empirical central limit theorems for level distributions in high-dimensional random fields, unifying diverse models under a Gaussian process framework.

Findings

Level distributions converge to a Gaussian process with regularly varying covariance.

Behavior differs from classical Brownian bridge limits for weakly dependent sequences.

Universal distribution pattern observed across multiple high-dimensional models.

Abstract

We prove empirical central limit theorems for the distribution of levels of various random fields defined on high-dimensional discrete structures as the dimension of the structure goes to $\infty$. The random fields considered include costs of assignments, weights of Hamiltonian cycles and spanning trees, energies of directed polymers, locations of particles in the branching random walk, as well as energies in the Sherrington--Kirkpatrick and Edwards--Anderson models. The distribution of levels in all models listed above is shown to be essentially the same as in a stationary Gaussian process with regularly varying nonsummable covariance function. This type of behavior is different from the Brownian bridge-type limit known for independent or stationary weakly dependent sequences of random variables.