Efficient Gradient-Free Variational Inference using Policy Search

Oleg Arenz, Gerhard Neumann, Mingjun Zhong
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:234-243, 2018.

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-arenz18a, title = {Efficient Gradient-Free Variational Inference using Policy Search}, author = {Arenz, Oleg and Neumann, Gerhard and Zhong, Mingjun}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {234--243}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/arenz18a/arenz18a.pdf}, url = {https://proceedings.mlr.press/v80/arenz18a.html}, 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.} }
Endnote
%0 Conference Paper %T Efficient Gradient-Free Variational Inference using Policy Search %A Oleg Arenz %A Gerhard Neumann %A Mingjun Zhong %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-arenz18a %I PMLR %P 234--243 %U https://proceedings.mlr.press/v80/arenz18a.html %V 80 %X 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.
APA
Arenz, O., Neumann, G. & Zhong, M.. (2018). Efficient Gradient-Free Variational Inference using Policy Search. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:234-243 Available from https://proceedings.mlr.press/v80/arenz18a.html.

Related Material