Susceptibility Propagation for Constraint Satisfaction Problems
Disordered Systems and Neural Networks
2010-07-29 v1 Statistical Mechanics
Information Theory
math.IT
Abstract
We study the susceptibility propagation, a message-passing algorithm to compute correlation functions. It is applied to constraint satisfaction problems and its accuracy is examined. As a heuristic method to find a satisfying assignment, we propose susceptibility-guided decimation where correlations among the variables play an important role. We apply this novel decimation to locked occupation problems, a class of hard constraint satisfaction problems exhibited recently. It is shown that the present method performs better than the standard belief-guided decimation.
Keywords
Cite
@article{arxiv.0903.1621,
title = {Susceptibility Propagation for Constraint Satisfaction Problems},
author = {Saburo Higuchi and Marc Mézard},
journal= {arXiv preprint arXiv:0903.1621},
year = {2010}
}
Comments
17 pages, 5 figures