Loop corrections for approximate inference

Abstract

We propose a method for improving approximate inference methods that corrects for the influence of loops in the graphical model. The method is applicable to arbitrary factor graphs, provided that the size of the Markov blankets is not too large. It is an alternative implementation of an idea introduced recently by Montanari and Rizzo (2005). In its simplest form, which amounts to the assumption that no loops are present, the method reduces to the minimal Cluster Variation Method approximation (which uses maximal factors as outer clusters). On the other hand, using estimates of the effect of loops (obtained by some approximate inference algorithm) and applying the Loop Correcting (LC) method usually gives significantly better results than applying the approximate inference algorithm directly without loop corrections. Indeed, we often observe that the loop corrected error is approximately the square of the error of the approximate inference method used to estimate the effect of loops. We compare different variants of the Loop Correcting method with other approximate inference methods on a variety of graphical models, including "real world" networks, and conclude that the LC approach generally obtains the most accurate results.

Technical report, 38 pages, 14 figures

References in corpus (6)

• Expectation Propagation for approximate Bayesian inference
• Residual Belief Propagation: Informed Scheduling for Asynchronous Message Passing
• Loop series for discrete statistical models on graphs
• Approximate Inference and Constrained Optimization
• Loop expansion around the Bethe-Peierls approximation for lattice models
• Multiplicative Factorization of Noisy-Max

Cited by in corpus (1)

• Truncating the loop series expansion for Belief Propagation