Convergent Propagation Algorithms via Oriented Trees

by   Amir Globerson, et al.

Inference problems in graphical models are often approximated by casting them as constrained optimization problems. Message passing algorithms, such as belief propagation, have previously been suggested as methods for solving these optimization problems. However, there are few convergence guarantees for such algorithms, and the algorithms are therefore not guaranteed to solve the corresponding optimization problem. Here we present an oriented tree decomposition algorithm that is guaranteed to converge to the global optimum of the Tree-Reweighted (TRW) variational problem. Our algorithm performs local updates in the convex dual of the TRW problem - an unconstrained generalized geometric program. Primal updates, also local, correspond to oriented reparametrization operations that leave the distribution intact.



page 1

page 2

page 3

page 4


Convex Combination Belief Propagation Algorithms

We introduce new message passing algorithms for inference with graphical...

Graphical Models for Optimal Power Flow

Optimal power flow (OPF) is the central optimization problem in electric...

On the Geometry of Message Passing Algorithms for Gaussian Reciprocal Processes

Reciprocal processes are acausal generalizations of Markov processes int...

Fundamental Linear Algebra Problem of Gaussian Inference

Underlying many Bayesian inference techniques that seek to approximate t...

Efficient MRF Energy Minimization via Adaptive Diminishing Smoothing

We consider the linear programming relaxation of an energy minimization ...

Variational Algorithms for Marginal MAP

Marginal MAP problems are notoriously difficult tasks for graphical mode...

Constrained fractional set programs and their application in local clustering and community detection

The (constrained) minimization of a ratio of set functions is a problem ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.