Neural Gaussian Copula for Variational Autoencoder

TL;DR

This paper introduces a neural Gaussian copula approach for variational autoencoders that explicitly models dependencies among latent variables, preventing posterior collapse and improving training stability.

Contribution

It proposes a novel Gaussian copula-based variational autoencoder that captures latent variable dependencies, addressing a key limitation of traditional VAEs.

Findings

Prevents posterior collapse in VAEs.

Achieves competitive results with explicit dependency modeling.

Improves training stability for variational language models.

Abstract

Variational language models seek to estimate the posterior of latent variables with an approximated variational posterior. The model often assumes the variational posterior to be factorized even when the true posterior is not. The learned variational posterior under this assumption does not capture the dependency relationships over latent variables. We argue that this would cause a typical training problem called posterior collapse observed in all other variational language models. We propose Gaussian Copula Variational Autoencoder (VAE) to avert this problem. Copula is widely used to model correlation and dependencies of high-dimensional random variables, and therefore it is helpful to maintain the dependency relationships that are lost in VAE. The empirical results show that by modeling the correlation of latent variables explicitly using a neural parametric copula, we can avert this training difficulty while getting competitive results among all other VAE approaches.