Efficient Gradient-Free Variational Inference using Policy Search

Oleg Arenz* (Corresponding Author), Gerhard Neumann, Mingjun Zhong

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Inference from complex distributions is a common problem in machine learning needed for many Bayesian methods. We propose an efficient, gradient-free method for learning general GMM approximations of multimodal distributions based on recent insights from stochastic search methods. Our method establishes information-geometric trust regions to ensure efficient exploration of the sampling space and stability of the GMM updates, allowing for efficient estimation of multi-variate Gaussian variational distributions. For GMMs, we apply a variational lower bound to decompose the learning objective into sub-problems given by learning the individual mixture components and the coefficients. The number of mixture components is adapted online in order to allow for arbitrary exact approximations. We demonstrate on several domains that we can learn significantly better approximations than competing variational inference methods and that the quality of samples drawn from our approximations is on par with samples created by state-of-the-art MCMC samplers that require significantly more computational resources.
Original languageEnglish
Title of host publicationProceedings of the 35th International Conference on Machine Learning
EditorsJennifer Dy, Andreas Krause
PublisherMLR Press
Pages234-243
Number of pages10
Volume80
Publication statusPublished - 1 Jul 2018
Event35th International Conference on Machine Learning, ICML 2018 - Stockholm, Sweden
Duration: 10 Jul 201815 Jul 2018

Publication series

NameProceedings of Machine Learning Research
ISSN (Electronic)2640-3498

Conference

Conference35th International Conference on Machine Learning, ICML 2018
CountrySweden
CityStockholm
Period10/07/1815/07/18

Fingerprint Dive into the research topics of 'Efficient Gradient-Free Variational Inference using Policy Search'. Together they form a unique fingerprint.

Cite this