Probability Functional Descent: A Unifying Perspective on GANs, Variational Inference, and Reinforcement Learning

Casey Chu, Jose Blanchet, Peter Glynn
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:1213-1222, 2019.

Abstract

The goal of this paper is to provide a unifying view of a wide range of problems of interest in machine learning by framing them as the minimization of functionals defined on the space of probability measures. In particular, we show that generative adversarial networks, variational inference, and actor-critic methods in reinforcement learning can all be seen through the lens of our framework. We then discuss a generic optimization algorithm for our formulation, called probability functional descent (PFD), and show how this algorithm recovers existing methods developed independently in the settings mentioned earlier.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-chu19a, title = {Probability Functional Descent: A Unifying Perspective on {GAN}s, Variational Inference, and Reinforcement Learning}, author = {Chu, Casey and Blanchet, Jose and Glynn, Peter}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {1213--1222}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/chu19a/chu19a.pdf}, url = {https://proceedings.mlr.press/v97/chu19a.html}, abstract = {The goal of this paper is to provide a unifying view of a wide range of problems of interest in machine learning by framing them as the minimization of functionals defined on the space of probability measures. In particular, we show that generative adversarial networks, variational inference, and actor-critic methods in reinforcement learning can all be seen through the lens of our framework. We then discuss a generic optimization algorithm for our formulation, called probability functional descent (PFD), and show how this algorithm recovers existing methods developed independently in the settings mentioned earlier.} }
Endnote
%0 Conference Paper %T Probability Functional Descent: A Unifying Perspective on GANs, Variational Inference, and Reinforcement Learning %A Casey Chu %A Jose Blanchet %A Peter Glynn %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-chu19a %I PMLR %P 1213--1222 %U https://proceedings.mlr.press/v97/chu19a.html %V 97 %X The goal of this paper is to provide a unifying view of a wide range of problems of interest in machine learning by framing them as the minimization of functionals defined on the space of probability measures. In particular, we show that generative adversarial networks, variational inference, and actor-critic methods in reinforcement learning can all be seen through the lens of our framework. We then discuss a generic optimization algorithm for our formulation, called probability functional descent (PFD), and show how this algorithm recovers existing methods developed independently in the settings mentioned earlier.
APA
Chu, C., Blanchet, J. & Glynn, P.. (2019). Probability Functional Descent: A Unifying Perspective on GANs, Variational Inference, and Reinforcement Learning. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:1213-1222 Available from https://proceedings.mlr.press/v97/chu19a.html.

Related Material