DMCA
Doubly stochastic variational Bayes for non-conjugate inference
| Citations: | 11 - 1 self |
BibTeX
@MISC{Titsias_doublystochastic,
author = {Michalis K. Titsias},
title = {Doubly stochastic variational Bayes for non-conjugate inference},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We propose a simple and effective variational inference algorithm based on stochastic optimi-sation that can be widely applied for Bayesian non-conjugate inference in continuous parameter spaces. This algorithm is based on stochastic ap-proximation and allows for efficient use of gra-dient information from the model joint density. We demonstrate these properties using illustra-tive examples as well as in challenging and di-verse Bayesian inference problems such as vari-able selection in logistic regression and fully Bayesian inference over kernel hyperparameters in Gaussian process regression. 1.
Keyphrases
stochastic variational bayes non-conjugate inference vari-able selection model joint density gaussian process regression kernel hyperparameters effective variational inference algorithm bayesian non-conjugate inference bayesian inference gra-dient information stochastic optimi-sation continuous parameter space logistic regression stochastic ap-proximation illustra-tive example di-verse bayesian inference problem efficient use
