Linear response algorithms for approximate inference in graphical models

  • Authors:
  • Max Welling;Yee Whye Teh

  • Affiliations:
  • Department of Computer Science, University of Toronto, Toronto M5S 3G4 Canada;Computer Science Division, University of California at Berkeley, Berkeley, CA

  • Venue:
  • Neural Computation
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Belief propagation (BP) on cyclic graphs is an efficient algorithm for computing approximate marginal probability distributions over single nodes and neighboring nodes in the graph. However, it does not prescribe a way to compute joint distributions over pairs of distant nodes in the graph. In this article, we propose two new algorithms for approximating these pairwise probabilities, based on the linear response theorem. The first is a propagation algorithm that is shown to converge if BP converges to a stable fixed point. The second algorithm is based on matrix inversion. Applying these ideas to gaussian random fields, we derive a propagation algorithm for computing the inverse of a matrix.