UPDF AI

Inadequacy of interval estimates corresponding to variational Bayesian approximations

Bo Wang,D. Titterington

2005 · DBLP: conf/aistats/0002T05
International Conference on Artificial Intelligence and Statistics · 125 Citations

TLDR

It is shown that the covariance matrices from the variational Bayes approximations are normally ‘too small’ compared with those for the maximum likelihood estimator, so that resulting interval estimates for the parameters will be unrealistically narrow.