PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Validity estimates for loopy belief propagation on binary real-world networks
Joris Mooij and Bert Kappen
In: Advances in Neural Information Processing Systems 17 (NIPS*2004) (2005) MIT Press , Cambridge, MA , pp. 945-952.

Abstract

We introduce a computationally efficient method to estimate the validity of the BP method as a function of graph topology, the connectivity strength, frustration and network size. We present numerical results that demonstrate the correctness of our estimates for the uniform random model and for a real-world network ("C. Elegans"). Although the method is restricted to pair-wise interactions, no local evidence (zero "biases") and binary variables, we believe that its predictions correctly capture the limitations of BP for inference and MAP estimation on arbitrary graphical models. Using this approach, we find that BP always performs better than MF. Especially for large networks with broad degree distributions (such as scale-free networks) BP turns out to significantly outperform MF.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
Postscript - Requires a viewer, such as GhostView
EPrint Type:Book Section
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:3218
Deposited By:Joris Mooij
Deposited On:20 January 2008