Probabilistic Arc Consistency: A Connection between Constraint Reasoning and Probabilistic Reasoning

Abstract

We document a connection between constraint reasoning and probabilistic reasoning. We present an algorithm, called {em probabilistic arc consistency}, which is both a generalization of a well known algorithm for arc consistency used in constraint reasoning, and a specialization of the belief updating algorithm for singly-connected networks. Our algorithm is exact for singly- connected constraint problems, but can work well as an approximation for arbitrary problems. We briefly discuss some empirical results, and related methods.