Survey propagation at finite temperature: application to a Sourlas code as a toy model
Disordered Systems and Neural Networks
2009-11-11 v1 Statistical Mechanics
Abstract
In this paper we investigate a finite temperature generalization of survey propagation, by applying it to the problem of finite temperature decoding of a biased finite connectivity Sourlas code for temperatures lower than the Nishimori temperature. We observe that the result is a shift of the location of the dynamical critical channel noise to larger values than the corresponding dynamical transition for belief propagation, as suggested recently by Migliorini and Saad for LDPC codes. We show how the finite temperature 1-RSB SP gives accurate results in the regime where competing approaches fail to converge or fail to recover the retrieval state.
Cite
@article{arxiv.cond-mat/0508586,
title = {Survey propagation at finite temperature: application to a Sourlas code as a toy model},
author = {B Wemmenhove and H J Kappen},
journal= {arXiv preprint arXiv:cond-mat/0508586},
year = {2009}
}