Improved Dynamic Schedules for Belief Propagation

TL;DR

This paper introduces an improved dynamic scheduling method for belief propagation that estimates message residuals to reduce computation and convergence time, achieving faster results without sacrificing accuracy.

Contribution

The paper proposes a novel residual estimation technique for dynamic message scheduling in belief propagation, significantly reducing message updates and runtime.

Findings

Reduces message count and runtime by up to five times.

Maintains solution quality while improving efficiency.

Effective on both synthetic and real-world networks.

Abstract

Belief propagation and its variants are popular methods for approximate inference, but their running time and even their convergence depend greatly on the schedule used to send the messages. Recently, dynamic update schedules have been shown to converge much faster on hard networks than static schedules, namely the residual BP schedule of Elidan et al. [2006]. But that RBP algorithm wastes message updates: many messages are computed solely to determine their priority, and are never actually performed. In this paper, we show that estimating the residual, rather than calculating it directly, leads to significant decreases in the number of messages required for convergence, and in the total running time. The residual is estimated using an upper bound based on recent work on message errors in BP. On both synthetic and real-world networks, this dramatically decreases the running time of BP, in some cases by a factor of five, without affecting the quality of the solution.