MIME: Mutual Information Minimization and Entropy Maximization for Bayesian Belief Propagation

Part of Advances in Neural Information Processing Systems 14 (NIPS 2001)

Bibtex Metadata Paper

Authors

Anand Rangarajan, Alan L. Yuille

Abstract

Bayesian belief propagation in graphical models has been recently shown to have very close ties to inference methods based in statis- tical physics. After Yedidia et al. demonstrated that belief prop- agation (cid:12)xed points correspond to extrema of the so-called Bethe free energy, Yuille derived a double loop algorithm that is guar- anteed to converge to a local minimum of the Bethe free energy. Yuille’s algorithm is based on a certain decomposition of the Bethe free energy and he mentions that other decompositions are possi- ble and may even be fruitful. In the present work, we begin with the Bethe free energy and show that it has a principled interpre- tation as pairwise mutual information minimization and marginal entropy maximization (MIME). Next, we construct a family of free energy functions from a spectrum of decompositions of the original Bethe free energy. For each free energy in this family, we develop a new algorithm that is guaranteed to converge to a local min- imum. Preliminary computer simulations are in agreement with this theoretical development.