Message passing in random satisfiability problems
Disordered Systems and Neural Networks
2007-05-23 v1 Statistical Mechanics
Abstract
This talk surveys the recent development of message passing procedures for solving constraint satisfaction problems. The cavity method from statistical physics provides a generalization of the belief propagation strategy that is able to deal with the clustering of solutions in these problems. It allows to derive analytic results on their phase diagrams, and offers a new algorithmic framework.
Keywords
Cite
@article{arxiv.cond-mat/0401237,
title = {Message passing in random satisfiability problems},
author = {Marc Mezard},
journal= {arXiv preprint arXiv:cond-mat/0401237},
year = {2007}
}
Comments
Talk given at NIPS 2003; 8 pages