English

Local approximate inference algorithms

Artificial Intelligence 2007-10-03 v3

Abstract

We present a new local approximation algorithm for computing Maximum a Posteriori (MAP) and log-partition function for arbitrary exponential family distribution represented by a finite-valued pair-wise Markov random field (MRF), say GG. Our algorithm is based on decomposition of GG into {\em appropriately} chosen small components; then computing estimates locally in each of these components and then producing a {\em good} global solution. We show that if the underlying graph GG either excludes some finite-sized graph as its minor (e.g. Planar graph) or has low doubling dimension (e.g. any graph with {\em geometry}), then our algorithm will produce solution for both questions within {\em arbitrary accuracy}. We present a message-passing implementation of our algorithm for MAP computation using self-avoiding walk of graph. In order to evaluate the computational cost of this implementation, we derive novel tight bounds on the size of self-avoiding walk tree for arbitrary graph. As a consequence of our algorithmic result, we show that the normalized log-partition function (also known as free-energy) for a class of {\em regular} MRFs will converge to a limit, that is computable to an arbitrary accuracy.

Keywords

Cite

@article{arxiv.cs/0610111,
  title  = {Local approximate inference algorithms},
  author = {Kyomin Jung and Devavrat Shah},
  journal= {arXiv preprint arXiv:cs/0610111},
  year   = {2007}
}

Comments

21 pages, 10 figures