Copulas as High-Dimensional Generative Models: Vine Copula Autoencoders

TL;DR

The paper introduces vine copula autoencoders (VCAEs), a flexible, low-cost method for high-dimensional data generation that leverages autoencoders and vine copulas to model complex distributions.

Contribution

It presents a novel three-step approach combining autoencoders and vine copulas to create a versatile generative model for high-dimensional data.

Findings

VCAEs achieve competitive results on benchmark datasets.

The method is computationally efficient compared to adversarial and variational models.

VCAEs can transform existing autoencoders into generative models.

Abstract

We introduce the vine copula autoencoder (VCAE), a flexible generative model for high-dimensional distributions built in a straightforward three-step procedure. First, an autoencoder (AE) compresses the data into a lower dimensional representation. Second, the multivariate distribution of the encoded data is estimated with vine copulas. Third, a generative model is obtained by combining the estimated distribution with the decoder part of the AE. As such, the proposed approach can transform any already trained AE into a flexible generative model at a low computational cost. This is an advantage over existing generative models such as adversarial networks and variational AEs which can be difficult to train and can impose strong assumptions on the latent space. Experiments on MNIST, Street View House Numbers and Large-Scale CelebFaces Attributes datasets show that VCAEs can achieve competitive results to standard baselines.