Variational approximations in Bayesian model selection for finite mixture distributions

Authors:
C. A. McGrory;D. M. Titterington
Affiliations:
University of Glasgow, UK;University of Glasgow, UK
Venue:
Computational Statistics & Data Analysis
Year:
2007

Citing 5
Cited 8

An Introduction to Variational Methods for Graphical Models

Machine Learning
Bayesian Learning for Neural Networks

Bayesian Learning for Neural Networks
Bayesian model search for mixture models based on optimizing variational bounds

Neural Networks
Variational Message Passing

The Journal of Machine Learning Research
Inferring parameters and structure of latent variable models by variational bayes

UAI'99 Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence

Editorial: Advances in Mixture Models

Computational Statistics & Data Analysis
Variational Bayes for estimating the parameters of a hidden Potts model

Statistics and Computing
Distributed variational Bayesian algorithms for Gaussian mixtures in sensor networks

Signal Processing
A knowledge-driven probabilistic framework for the prediction of protein-protein interaction networks

Computers in Biology and Medicine
Online kernel density estimation for interactive learning

Image and Vision Computing
A new variational Bayesian algorithm with application to human mobility pattern modeling

Statistics and Computing
Initializing the EM algorithm in Gaussian mixture models with an unknown number of components

Computational Statistics & Data Analysis
Bayesian binary regression with exponential power link

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.03

Visualization

Abstract

Variational methods, which have become popular in the neural computing/machine learning literature, are applied to the Bayesian analysis of mixtures of Gaussian distributions. It is also shown how the deviance information criterion, (DIC), can be extended to these types of model by exploiting the use of variational approximations. The use of variational methods for model selection and the calculation of a DIC are illustrated with real and simulated data. The variational approach allows the simultaneous estimation of the component parameters and the model complexity. It is found that initial selection of a large number of components results in superfluous components being eliminated as the method converges to a solution. This corresponds to an automatic choice of model complexity. The appropriateness of this is reflected in the DIC values.