Scaling MPE Inference for Constrained Continuous Markov Random Fields with Consensus Optimization

TitleScaling MPE Inference for Constrained Continuous Markov Random Fields with Consensus Optimization
Publication TypeConference Paper
Year of Publication2012
AuthorsBach, S, Broecheler, M, Getoor, L, O'Leary, D
Conference NameNeuRIPS
Abstract

Probabilistic graphical models are powerful tools for analyzing constrained, continuous domains. However, finding most-probable explanations (MPEs) in these models can be computationally expensive. In this paper, we improve the scalability of MPE inference in a class of graphical models with piecewise-linear and piecewise-quadratic dependencies and linear constraints over continuous domains. We derive algorithms based on a consensus-optimization framework and demonstrate their superior performance over state of the art. We show empirically that in a large-scale voter-preference modeling problem our algorithms scale linearly in the number of dependencies and constraints