English

Bethe-ADMM for Tree Decomposition based Parallel MAP Inference

Artificial Intelligence 2013-09-27 v1 Machine Learning Machine Learning

Abstract

We consider the problem of maximum a posteriori (MAP) inference in discrete graphical models. We present a parallel MAP inference algorithm called Bethe-ADMM based on two ideas: tree-decomposition of the graph and the alternating direction method of multipliers (ADMM). However, unlike the standard ADMM, we use an inexact ADMM augmented with a Bethe-divergence based proximal function, which makes each subproblem in ADMM easy to solve in parallel using the sum-product algorithm. We rigorously prove global convergence of Bethe-ADMM. The proposed algorithm is extensively evaluated on both synthetic and real datasets to illustrate its effectiveness. Further, the parallel Bethe-ADMM is shown to scale almost linearly with increasing number of cores.

Keywords

Cite

@article{arxiv.1309.6829,
  title  = {Bethe-ADMM for Tree Decomposition based Parallel MAP Inference},
  author = {Qiang Fu and Huahua Wang and Arindam Banerjee},
  journal= {arXiv preprint arXiv:1309.6829},
  year   = {2013}
}

Comments

Appears in Proceedings of the Twenty-Ninth Conference on Uncertainty in Artificial Intelligence (UAI2013)

R2 v1 2026-06-22T01:34:32.552Z