Variational Learning Based on Infinite Inverted Dirichlet Mixture Model
WANG Jingzhong
YU Pengpeng
Abstract：Recent studies have shown that finite inverse Dirichlet mixture model is an important model for modeling non Gauss data.However, it has the problem of parameter estimation and model selection.The EM algorithm can not be used to accurately estimate the parameters and select the optimal number of mixture components.Therefore, this paper studies the infinite inverse Dirichlet mixture model, presents a novel variational approximate inference algorithm for learning.The algorithm can solve these two problems at the same time.In order to verify the effectiveness of the algorithm, this paper carries out experiments on artificial data sets.Experimental results show that the variational Bayesian inference to estimate mixed infinite inverse Dirichlet distribution is a very effective method.
Keywords：inverted Dirichletvariational inferenceBayesian estimationparameter estimationmodel selection
Publication Date：2017-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 640-644 )