Density Propagation and Improved Bounds on the Partition Function

Given a probabilistic graphical model, its density of states is a distribution that, for any likelihood value, gives the number of configurations with that probability. We introduce a novel message-passing algorithm called Density Propagation (DP) for estimating this distribution. We show that DP is exact for tree-structured graphical models and is, in general, a strict generalization of both sum-product and max-product algorithms. Further, we use density of states and tree decomposition to introduce a new family of upper and lower bounds on the partition function. For any tree decomposition, the new upper bound based on finer-grained density of state information is provably at least as tight as previously known bounds based on convexity of the log-partition function, and strictly stronger if a general condition holds. We conclude with empirical evidence of improvement over convex relaxations and mean-field based bounds.

[1]  G. Parisi,et al.  Statistical Field Theory , 1988 .

[2]  W. Cheung,et al.  Generalizations of Hölder's inequality , 2001 .

[3]  Michael I. Jordan,et al.  Graphical Models, Exponential Families, and Variational Inference , 2008, Found. Trends Mach. Learn..

[4]  M. Powell.  The BOBYQA algorithm for bound constrained optimization without derivatives , 2009 .

[5]  Martin J. Wainwright,et al.  Stochastic processes on graphs with cycles: geometric and variational approaches , 2002 .

[6]  John W. Fisher,et al.  Loopy Belief Propagation: Convergence and Effects of Message Errors , 2005, J. Mach. Learn. Res..

[7]  Gerhard Lakemeyer,et al.  Exploring artificial intelligence in the new millennium , 2003 .

[8]  Robert J. McEliece,et al.  The generalized distributive law , 2000, IEEE Trans. Inf. Theory.

[9]  Yair Weiss,et al.  MAP Estimation, Linear Programming and Belief Propagation with Convex Free Energies , 2007, UAI.

[10]  Qiang Liu,et al.  Negative Tree Reweighted Belief Propagation , 2010, UAI.

[11]  Matthew Richardson,et al.  Markov logic networks , 2006, Machine Learning.

[12]  Joris M. Mooij,et al.  libDAI: A Free and Open Source C++ Library for Discrete Approximate Inference in Graphical Models , 2010, J. Mach. Learn. Res..

[13]  D. Landau,et al.  Efficient, multiple-range random walk algorithm to calculate the density of states. , 2000, Physical review letters.

[14]  William T. Freeman,et al.  Understanding belief propagation and its generalizations , 2003 .

[15]  Lawrence D. Brown.  Fundamentals of Statistical Exponential Families , 1987 .

[16]  Michael I. Jordan.  Graphical Models , 1998 .

[17]  Martin J. Wainwright,et al.  Exact MAP Estimates by (Hyper)tree Agreement , 2002, NIPS.

[18]  Hilbert J. Kappen,et al.  Sufficient Conditions for Convergence of the Sum–Product Algorithm , 2005, IEEE Transactions on Information Theory.

[19]  Martin J. Wainwright,et al.  Tree-reweighted belief propagation algorithms and approximate ML estimation by pseudo-moment matching , 2003, AISTATS.

[20]  Amir Globerson,et al.  Convergent message passing algorithms - a unifying view , 2009, UAI.

[21]  Bart Selman,et al.  Accelerated Adaptive Markov Chain for Partition Function Computation , 2011, NIPS.

[22]  Tommi S. Jaakkola,et al.  Tightening LP Relaxations for MAP using Message Passing , 2008, UAI.

[23]  Michael I. Jordan,et al.  Loopy Belief Propagation for Approximate Inference: An Empirical Study , 1999, UAI.

[24]  Tamir Hazan,et al.  Norm-Product Belief Propagation: Primal-Dual Message-Passing for Approximate Inference , 2009, IEEE Transactions on Information Theory.