Fast Variational Inference in the Conjugate Exponential Family
Abstract
We present a general method for deriving collapsed variational inference algorithms for probabilistic models in the conjugate exponential family. Our method unifies many existing approaches to collapsed variational inference. Our collapsed variational inference leads to a new lower bound on the marginal likelihood. We exploit the information geometry of the bound to derive much faster optimization methods based on conjugate gradients for these models. Our approach is very general and is easily applied to any model where the mean field update equations have been derived. Empirically we show significant speed-ups for probabilistic models optimized using our bound.
Cite
Text
Hensman et al. "Fast Variational Inference in the Conjugate Exponential Family." Neural Information Processing Systems, 2012.Markdown
[Hensman et al. "Fast Variational Inference in the Conjugate Exponential Family." Neural Information Processing Systems, 2012.](https://mlanthology.org/neurips/2012/hensman2012neurips-fast/)BibTeX
@inproceedings{hensman2012neurips-fast,
title = {{Fast Variational Inference in the Conjugate Exponential Family}},
author = {Hensman, James and Rattray, Magnus and Lawrence, Neil D.},
booktitle = {Neural Information Processing Systems},
year = {2012},
pages = {2888-2896},
url = {https://mlanthology.org/neurips/2012/hensman2012neurips-fast/}
}