Convergence and asymptotic normality of variational Bayesian approximations for exponential family models with missing values
Statistics Theory
2012-07-19 v1 Computation
Methodology
Statistics Theory
Abstract
We study the properties of variational Bayes approximations for exponential family models with missing values. It is shown that the iterative algorithm for obtaining the variational Bayesian estimator converges locally to the true value with probability 1 as the sample size becomes inde nitely large. Moreover, the variational posterior distribution is proved to be asymptotically normal.
Keywords
Cite
@article{arxiv.1207.4159,
title = {Convergence and asymptotic normality of variational Bayesian approximations for exponential family models with missing values},
author = {Bo Wang and D. Titterington},
journal= {arXiv preprint arXiv:1207.4159},
year = {2012}
}
Comments
Appears in Proceedings of the Twentieth Conference on Uncertainty in Artificial Intelligence (UAI2004)