Title :
Parameter estimation of dependence tree models using the EM algorithm
Author :
Ronen, O. ; Rohlicek, J.R. ; Ostendorf, M.
Author_Institution :
Coll. of Eng., Boston Univ., MA, USA
Abstract :
A dependence tree is a model for the joint probability distribution of an n-dimensional random vector, which requires a relatively small number of free parameters by making Markov-like assumptions on the tree. The authors address the problem of maximum likelihood estimation of dependence tree models with missing observations, using the expectation-maximization algorithm. The solution involves computing observation probabilities with an iterative "upward-downward" algorithm, which is similar to an algorithm proposed for belief propagation in causal trees, a special case of Bayesian networks.<>
Keywords :
Markov processes; iterative methods; maximum likelihood estimation; optimisation; probability; random processes; signal processing; tree data structures; Bayesian networks; EM algorithm; Markov-like assumptions; dependence tree models; expectation-maximization algorithm; iterative upward-downward algorithm; joint probability distribution; maximum likelihood estimation; missing observations; n-dimensional random vector; observation probabilities; parameter estimation; Bayesian methods; Belief propagation; Computer networks; Iterative algorithms; Joining processes; Maximum likelihood estimation; Parameter estimation; Probability distribution; Signal processing algorithms; Topology;
Journal_Title :
Signal Processing Letters, IEEE
