Accelerated Message Passing for Entropy-Regularized MAP Inference

TL;DR

This paper introduces accelerated randomized message passing algorithms for entropy-regularized MAP inference in Markov random fields, achieving faster convergence and more efficient recovery of MAP solutions.

Contribution

It develops novel accelerated variants of smooth message passing algorithms using randomized techniques, improving convergence rates for entropy-regularized MAP inference.

Findings

Accelerated algorithms converge faster to ε-optimal solutions.

Fewer iterations needed to recover true MAP when LP is tight.

Proven theoretical convergence guarantees for the proposed methods.

Abstract

Maximum a posteriori (MAP) inference in discrete-valued Markov random fields is a fundamental problem in machine learning that involves identifying the most likely configuration of random variables given a distribution. Due to the difficulty of this combinatorial problem, linear programming (LP) relaxations are commonly used to derive specialized message passing algorithms that are often interpreted as coordinate descent on the dual LP. To achieve more desirable computational properties, a number of methods regularize the LP with an entropy term, leading to a class of smooth message passing algorithms with convergence guarantees. In this paper, we present randomized methods for accelerating these algorithms by leveraging techniques that underlie classical accelerated gradient methods. The proposed algorithms incorporate the familiar steps of standard smooth message passing algorithms, which can be viewed as coordinate minimization steps. We show that these accelerated variants achieve faster rates for finding $\epsilon$-optimal points of the unregularized problem, and, when the LP is tight, we prove that the proposed algorithms recover the true MAP solution in fewer iterations than standard message passing algorithms.