Complexity of Bethe Approximation

Jinwoo Shin
; Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, PMLR 22:1037-1045, 2012.

Abstract

This paper resolves a common complexity issue in the Bethe approximation of statistical physics and the sum-product Belief Propagation (BP) algorithm of artificial intelligence. The Bethe approximation reduces the problem of computing the partition function in a graphical model to that of solving a set of non-linear equations, so-called the Bethe equation. On the other hand, the BP algorithm is a popular heuristic method for estimating marginal distribution in a graphical model. Although they are inspired and developed from different directions, Yedidia, Freeman and Weiss (2004) established a somewhat surprising connection: the BP algorithm solves the Bethe equation if it converges (however, it often does not). This naturally motivates the following important question to understand their limitations and empirical successes: the Bethe equation is computationally easy to solve? We present a message passing algorithm solving the Bethe equation in polynomial number of bitwise operations for arbitrary binary graphical models of n nodes where the maximum degree in the underlying graph is O(log n). Our algorithm, an alternative to BP fixing its convergence issue, is the first fully polynomial-time approximation scheme for the BP fixed point computation in such a large class of graphical models. Moreover, we believe that our technique is of broader interest to understand the computational complexity of the cavity method in statistical physics.

Cite this Paper


BibTeX
@InProceedings{pmlr-v22-shin12, title = {Complexity of Bethe Approximation}, author = {Jinwoo Shin}, booktitle = {Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics}, pages = {1037--1045}, year = {2012}, editor = {Neil D. Lawrence and Mark Girolami}, volume = {22}, series = {Proceedings of Machine Learning Research}, address = {La Palma, Canary Islands}, month = {21--23 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v22/shin12/shin12.pdf}, url = {http://proceedings.mlr.press/v22/shin12.html}, abstract = {This paper resolves a common complexity issue in the Bethe approximation of statistical physics and the sum-product Belief Propagation (BP) algorithm of artificial intelligence. The Bethe approximation reduces the problem of computing the partition function in a graphical model to that of solving a set of non-linear equations, so-called the Bethe equation. On the other hand, the BP algorithm is a popular heuristic method for estimating marginal distribution in a graphical model. Although they are inspired and developed from different directions, Yedidia, Freeman and Weiss (2004) established a somewhat surprising connection: the BP algorithm solves the Bethe equation if it converges (however, it often does not). This naturally motivates the following important question to understand their limitations and empirical successes: the Bethe equation is computationally easy to solve? We present a message passing algorithm solving the Bethe equation in polynomial number of bitwise operations for arbitrary binary graphical models of n nodes where the maximum degree in the underlying graph is O(log n). Our algorithm, an alternative to BP fixing its convergence issue, is the first fully polynomial-time approximation scheme for the BP fixed point computation in such a large class of graphical models. Moreover, we believe that our technique is of broader interest to understand the computational complexity of the cavity method in statistical physics.} }
Endnote
%0 Conference Paper %T Complexity of Bethe Approximation %A Jinwoo Shin %B Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2012 %E Neil D. Lawrence %E Mark Girolami %F pmlr-v22-shin12 %I PMLR %J Proceedings of Machine Learning Research %P 1037--1045 %U http://proceedings.mlr.press %V 22 %W PMLR %X This paper resolves a common complexity issue in the Bethe approximation of statistical physics and the sum-product Belief Propagation (BP) algorithm of artificial intelligence. The Bethe approximation reduces the problem of computing the partition function in a graphical model to that of solving a set of non-linear equations, so-called the Bethe equation. On the other hand, the BP algorithm is a popular heuristic method for estimating marginal distribution in a graphical model. Although they are inspired and developed from different directions, Yedidia, Freeman and Weiss (2004) established a somewhat surprising connection: the BP algorithm solves the Bethe equation if it converges (however, it often does not). This naturally motivates the following important question to understand their limitations and empirical successes: the Bethe equation is computationally easy to solve? We present a message passing algorithm solving the Bethe equation in polynomial number of bitwise operations for arbitrary binary graphical models of n nodes where the maximum degree in the underlying graph is O(log n). Our algorithm, an alternative to BP fixing its convergence issue, is the first fully polynomial-time approximation scheme for the BP fixed point computation in such a large class of graphical models. Moreover, we believe that our technique is of broader interest to understand the computational complexity of the cavity method in statistical physics.
RIS
TY - CPAPER TI - Complexity of Bethe Approximation AU - Jinwoo Shin BT - Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics PY - 2012/03/21 DA - 2012/03/21 ED - Neil D. Lawrence ED - Mark Girolami ID - pmlr-v22-shin12 PB - PMLR SP - 1037 DP - PMLR EP - 1045 L1 - http://proceedings.mlr.press/v22/shin12/shin12.pdf UR - http://proceedings.mlr.press/v22/shin12.html AB - This paper resolves a common complexity issue in the Bethe approximation of statistical physics and the sum-product Belief Propagation (BP) algorithm of artificial intelligence. The Bethe approximation reduces the problem of computing the partition function in a graphical model to that of solving a set of non-linear equations, so-called the Bethe equation. On the other hand, the BP algorithm is a popular heuristic method for estimating marginal distribution in a graphical model. Although they are inspired and developed from different directions, Yedidia, Freeman and Weiss (2004) established a somewhat surprising connection: the BP algorithm solves the Bethe equation if it converges (however, it often does not). This naturally motivates the following important question to understand their limitations and empirical successes: the Bethe equation is computationally easy to solve? We present a message passing algorithm solving the Bethe equation in polynomial number of bitwise operations for arbitrary binary graphical models of n nodes where the maximum degree in the underlying graph is O(log n). Our algorithm, an alternative to BP fixing its convergence issue, is the first fully polynomial-time approximation scheme for the BP fixed point computation in such a large class of graphical models. Moreover, we believe that our technique is of broader interest to understand the computational complexity of the cavity method in statistical physics. ER -
APA
Shin, J.. (2012). Complexity of Bethe Approximation. Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, in PMLR 22:1037-1045

Related Material