English

Convexifying the Bethe Free Energy

Artificial Intelligence 2012-05-14 v1 Machine Learning

Abstract

The introduction of loopy belief propagation (LBP) revitalized the application of graphical models in many domains. Many recent works present improvements on the basic LBP algorithm in an attempt to overcome convergence and local optima problems. Notable among these are convexified free energy approximations that lead to inference procedures with provable convergence and quality properties. However, empirically LBP still outperforms most of its convex variants in a variety of settings, as we also demonstrate here. Motivated by this fact we seek convexified free energies that directly approximate the Bethe free energy. We show that the proposed approximations compare favorably with state-of-the art convex free energy approximations.

Keywords

Cite

@article{arxiv.1205.2624,
  title  = {Convexifying the Bethe Free Energy},
  author = {Ofer Meshi and Ariel Jaimovich and Amir Globerson and Nir Friedman},
  journal= {arXiv preprint arXiv:1205.2624},
  year   = {2012}
}

Comments

Appears in Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence (UAI2009)

R2 v1 2026-06-21T21:02:29.611Z