Computer Science – Learning
Scientific paper
2012-03-15
Computer Science
Learning
Appears in Proceedings of the Twenty-Sixth Conference on Uncertainty in Artificial Intelligence (UAI2010)
Scientific paper
It is known that fixed points of loopy belief propagation (BP) correspond to stationary points of the Bethe variational problem, where we minimize the Bethe free energy subject to normalization and marginalization constraints. Unfortunately, this does not entirely explain BP because BP is a dual rather than primal algorithm to solve the Bethe variational problem -- beliefs are infeasible before convergence. Thus, we have no better understanding of BP than as an algorithm to seek for a common zero of a system of non-linear functions, not explicitly related to each other. In this theoretical paper, we show that these functions are in fact explicitly related -- they are the partial derivatives of a single function of reparameterizations. That means, BP seeks for a stationary point of a single function, without any constraints. This function has a very natural form: it is a linear combination of local log-partition functions, exactly as the Bethe entropy is the same linear combination of local entropies.
No associations
LandOfFree
Primal View on Belief Propagation does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Primal View on Belief Propagation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Primal View on Belief Propagation will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-32387