Efficient Determination of Gibbs Estimators with Submodular Energy Functions
Optimization and Control
2007-06-13 v1 Statistics Theory
Statistics Theory
Abstract
Combinatorial algorithms for minimization of functions of many variables, which take their values in finite totally ordered sets, are developed. For that the decomposition of the functions by Boolean polynomials is used. The modified SFM algorithm for minimization of submodular Boolean polynomials is described. The representation of some subclass of submodular Boolean polynomials by graphs is derived in order to make possible use of the graph cut technique.
Cite
@article{arxiv.math/0304041,
title = {Efficient Determination of Gibbs Estimators with Submodular Energy Functions},
author = {Boris Zalesky},
journal= {arXiv preprint arXiv:math/0304041},
year = {2007}
}
Comments
18 pages, 3 figures