Title :
1|Functions on probabilistic graphical models
Author :
Ignac, Tomasz ; Sorger, Uli
Author_Institution :
Fac. of Sci., Technol. & Commun., Univ. of Luxembourg, Luxembourg, Luxembourg
Abstract :
Probabilistic graphical models are tools that are used to represent the probability distribution of a vector of random variables X = (X1, ..., XN). In this paper we introduce functions f(x1, ..., xN) defined over the given vector. These functions also are random variables. The main result of the paper is an algorithm for finding the expected value and other moments for some classes of f(x1, ..., xN). The possible applications of that algorithm are discussed. Specifically, we use it to analyze the entropy of X and to compute the relative entropy of two probability distributions of the same vector X. Finally, open problems and possible topics of future researches are discussed.
Keywords :
entropy; graph theory; statistical distributions; probabilistic graphical models; probability distribution; random variables; relative entropy; Computer science; Decision making; Distributed computing; Entropy; Graphical models; Information technology; Markov random fields; Probability distribution; Random variables; Uncertainty;
Conference_Titel :
Computer Science and Information Technology, 2009. IMCSIT '09. International Multiconference on
Conference_Location :
Mragowo
Print_ISBN :
978-1-4244-5314-6
DOI :
10.1109/IMCSIT.2009.5352794