VMI-VAE: Variational Mutual Information Maximization Framework for VAE With Discrete and Continuous Priors
This addresses representation quality evaluation in VAEs, but it appears incremental as it builds on existing VAE methods.
The authors tackled the problem of VAE not explicitly measuring representation quality by proposing a Variational Mutual Information Maximization Framework, which forces VAE to not ignore latent codes and allows selection of informative components, though no concrete performance numbers are provided.
Variational Autoencoder is a scalable method for learning latent variable models of complex data. It employs a clear objective that can be easily optimized. However, it does not explicitly measure the quality of learned representations. We propose a Variational Mutual Information Maximization Framework for VAE to address this issue. It provides an objective that maximizes the mutual information between latent codes and observations. The objective acts as a regularizer that forces VAE to not ignore the latent code and allows one to select particular components of it to be most informative with respect to the observations. On top of that, the proposed framework provides a way to evaluate mutual information between latent codes and observations for a fixed VAE model.