Factorization of Discrete Probability Distributions
Artificial Intelligence
2013-01-07 v1
Abstract
We formulate necessary and sufficient conditions for an arbitrary discrete probability distribution to factor according to an undirected graphical model, or a log-linear model, or other more general exponential models. This result generalizes the well known Hammersley-Clifford Theorem.
Cite
@article{arxiv.1301.0568,
title = {Factorization of Discrete Probability Distributions},
author = {Dan Geiger and Christopher Meek and Bernd Sturmfels},
journal= {arXiv preprint arXiv:1301.0568},
year = {2013}
}
Comments
Appears in Proceedings of the Eighteenth Conference on Uncertainty in Artificial Intelligence (UAI2002)