Statistics and Learning Theory
L.10.3
Lecture
Learning Theory I

Learning Representations by Maximizing Mutual Information in Variational Autoencoders

Sriram Vishwanath, Ali Lotfi Rezaabad

Date & Time

01:00 am – 01:00 am

Abstract

Variational autoencoders (VAE) have ushered in an new era of unsupervised learning methods for complex distributions. Although these techniques are elegant in their approach, they are typically not useful for representation learning. In this work, we propose a simple yet powerful class of VAEs that simultaneously result in meaningful learned representations. Our solution is to combine traditional VAEs with mutual information maximization, with the goal to enhance amortized inference in VAEs using Information Theoretic techniques. We call this approach InfoMax-VAE, and such an approach can significantly boost the quality of learned high-level representations. We realize this through explicit maximization of information measures associated with the representation. Using extensive experiments on varied datasets and setups, we show that InfoMax-VAE outperforms contemporary popular approaches, including Info-VAE and $\beta$-VAE.


Presenters

Sriram Vishwanath

The University of Texas at Austin

Ali Lotfi Rezaabad

The University of Texas at Austin

Session Chair

Lizhong Zheng

Massachusetts Institute of Technology