Mean-field approximation, convex hierarchies, and the optimality of correlation rounding: a unified perspective

TL;DR

This paper unifies mean-field and convex hierarchy methods for approximating the Ising model's free energy, showing their tight regimes and introducing algorithms with provable guarantees, while also establishing the optimality of correlation rounding.

Contribution

It demonstrates the equivalence of tight regimes for mean-field and convex relaxations in free energy approximation, introduces a simple algorithmic proof, and proves the optimality of correlation rounding.

Findings

Mean-field approximation is within $O((n\|J\\|_{F})^{2/3})$ of the free energy.

The tight regime for both methods is identical, subsuming previous results.

Correlation rounding is proven to be optimal, refuting a recent conjecture.

Abstract

The free energy is a key quantity of interest in Ising models, but unfortunately, computing it in general is computationally intractable. Two popular (variational) approximation schemes for estimating the free energy of general Ising models (in particular, even in regimes where correlation decay does not hold) are: (i) the mean-field approximation with roots in statistical physics, which estimates the free energy from below, and (ii) hierarchies of convex relaxations with roots in theoretical computer science, which estimate the free energy from above. We show, surprisingly, that the tight regime for both methods to compute the free energy to leading order is identical. More precisely, we show that the mean-field approximation is within $O((n\|J\|_{F})^{2/3})$ of the free energy, where $\|J\|_F$ denotes the Frobenius norm of the interaction matrix of the Ising model. This simultaneously subsumes both the breakthrough work of Basak and Mukherjee, who showed the tight result that the mean-field approximation is within $o(n)$ whenever $\|J\|_{F} = o(\sqrt{n})$, as well as the work of Jain, Koehler, and Mossel, who gave the previously best known non-asymptotic bound of $O((n\|J\|_{F})^{2/3}\log^{1/3}(n\|J\|_{F}))$. We give a simple, algorithmic proof of this result using a convex relaxation proposed by Risteski based on the Sherali-Adams hierarchy, automatically giving sub-exponential time approximation schemes for the free energy in this entire regime. Our algorithmic result is tight under Gap-ETH. We furthermore combine our techniques with spin glass theory to prove (in a strong sense) the optimality of correlation rounding, refuting a recent conjecture of Allen, O'Donnell, and Zhou. Finally, we give the tight generalization of all of these results to $k$-MRFs, capturing as a special case previous work on approximating MAX-$k$-CSP.