DocumentCode :
3120262
Title :
The Bethe approximation of the pattern maximum likelihood distribution
Author :
Vontobel, Pascal O.
Author_Institution :
Hewlett-Packard Labs., Palo Alto, CA, USA
fYear :
2012
fDate :
1-6 July 2012
Firstpage :
2012
Lastpage :
2016
Abstract :
Among all memoryless source distributions, the pattern maximum likelihood (PML) distribution is the distribution which maximizes the probability that a memoryless source produces a string with a given pattern. Equivalently, the PML distribution maximizes the permanent of a certain non-negative matrix. We reformulate this maximization problem as a double minimization problem of a suitable Gibbs free energy function. Because finding the minimum of this function appears intractable for practically relevant problem sizes, one must look for tractable approximations. One approach is to approximately find a minimum (or at least a local minimum) of the Gibbs free energy function by applying an alternating minimization algorithm where the steps are based on quantities that are obtained by Markov chain Monte Carlo sampling. One can show that this approach is equivalent to an algorithm that was proposed by Orlitsky et al. An alternative approach is to replace the Gibbs free energy function by a tractable approximation like the Bethe free energy function and to apply an alternating minimization algorithm to this function. As it turns out, empirically, this latter approach gives very good approximations to the PML distribution (or at least a locally optimal PML distribution), and, for the same level of accuracy, is two to three orders of magnitude faster than the former approach for practically relevant problem sizes. Moreover, the above free energy framework allows us to simplify some earlier proofs of properties of the PML distribution and to derive some new properties of the PML distribution, along with obtaining similar results for its Bethe approximation.
Keywords :
Markov processes; Monte Carlo methods; approximation theory; maximum likelihood estimation; minimisation; sampling methods; statistical distributions; Bethe approximation; Bethe free energy function; Gibbs free energy function; Markov chain Monte Carlo sampling; alternating minimization algorithm; double minimization problem; maximization problem; memoryless source distribution; nonnegative matrix; pattern maximum likelihood distribution; probability; tractable approximation; Approximation algorithms; Approximation methods; Graphical models; Maximum likelihood estimation; Mercury (metals); Minimization; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Theory Proceedings (ISIT), 2012 IEEE International Symposium on
Conference_Location :
Cambridge, MA
ISSN :
2157-8095
Print_ISBN :
978-1-4673-2580-6
Electronic_ISBN :
2157-8095
Type :
conf
DOI :
10.1109/ISIT.2012.6283654
Filename :
6283654
Link To Document :
بازگشت