Abstract:
Monte Carlo methods are considered to compute, first, the partition function of graphical models, and second, the information rate of source/channel models, in both cases...Show MoreMetadata
Abstract:
Monte Carlo methods are considered to compute, first, the partition function of graphical models, and second, the information rate of source/channel models, in both cases for factor graphs with cycles. The convergence of two basic Monte Carlo methods is improved by sampling only a cycle breaking subset of the variables and using exact sum-product computations for the remaining variables. The methods are demonstrated by their application to a two-dimensional Ising model and to a two-dimensional intersymbol interference channel.
Published in: 2008 IEEE International Symposium on Information Theory
Date of Conference: 06-11 July 2008
Date Added to IEEE Xplore: 08 August 2008
ISBN Information: