ML Anthology

Convergence and Asymptotic Normality of Variational Bayesian Approximations for Expon

Bo Wang, D. M. Titterington
UAI 2004 pp. 577-584
/uai/2004/wang2004uai-convergence/

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 indefinitely large. Moreover, the variational posterior distribution is proved to be asymptotically normal.

PDF UAI Semantic Scholar

Cite

Text

Wang and Titterington. "Convergence and Asymptotic Normality of Variational Bayesian Approximations for Expon." Conference on Uncertainty in Artificial Intelligence, 2004.

Markdown

[Wang and Titterington. "Convergence and Asymptotic Normality of Variational Bayesian Approximations for Expon." Conference on Uncertainty in Artificial Intelligence, 2004.](https://mlanthology.org/uai/2004/wang2004uai-convergence/)

BibTeX

@inproceedings{wang2004uai-convergence,
  title     = {{Convergence and Asymptotic Normality of Variational Bayesian Approximations for Expon}},
  author    = {Wang, Bo and Titterington, D. M.},
  booktitle = {Conference on Uncertainty in Artificial Intelligence},
  year      = {2004},
  pages     = {577-584},
  url       = {https://mlanthology.org/uai/2004/wang2004uai-convergence/}
}