Title :
Consistent estimation of the basic neighborhood of Markov random fields
Author :
Csiszár, Imre ; Talata, Zsolt
Author_Institution :
A. Renyi Inst. of Math., Hungarian Acad. of Sci., Hungary
fDate :
27 June-2 July 2004
Abstract :
For Markov random fields with finite set of states, a modification of the Bayesian information criterion admits strongly consistent estimation of the smallest region that determines the conditional distributions. Phase transition or nonstationarity do not affect the result.
Keywords :
Bayes methods; Markov processes; maximum likelihood estimation; Bayesian information criterion; Markov random field; conditional distribution; consistent estimation; phase transition; Bayesian methods; Markov random fields; Mathematics; Maximum likelihood estimation; Random variables; State estimation;
Conference_Titel :
Information Theory, 2004. ISIT 2004. Proceedings. International Symposium on
Print_ISBN :
0-7803-8280-3
DOI :
10.1109/ISIT.2004.1365204
