NCP-VAE: Variational Autoencoders with Noise Contrastive Priors

by   Jyoti Aneja, et al.

Variational autoencoders (VAEs) are one of the powerful likelihood-based generative models with applications in various domains. However, they struggle to generate high-quality images, especially when samples are obtained from the prior without any tempering. One explanation for VAEs' poor generative quality is the prior hole problem: the prior distribution fails to match the aggregate approximate posterior. Due to this mismatch, there exist areas in the latent space with high density under the prior that do not correspond to any encoded image. Samples from those areas are decoded to corrupted images. To tackle this issue, we propose an energy-based prior defined by the product of a base prior distribution and a reweighting factor, designed to bring the base closer to the aggregate posterior. We train the reweighting factor by noise contrastive estimation, and we generalize it to hierarchical VAEs with many latent variable groups. Our experiments confirm that the proposed noise contrastive priors improve the generative performance of state-of-the-art VAEs by a large margin on the MNIST, CIFAR-10, CelebA 64, and CelebA HQ 256 datasets.


page 6

page 18

page 19

page 20

page 21

page 22


Perceptual Generative Autoencoders

Modern generative models are usually designed to match target distributi...

LDC-VAE: A Latent Distribution Consistency Approach to Variational AutoEncoders

Variational autoencoders (VAEs), as an important aspect of generative mo...

Generalizing Variational Autoencoders with Hierarchical Empirical Bayes

Variational Autoencoders (VAEs) have experienced recent success as data-...

Fully Variational Noise-Contrastive Estimation

By using the underlying theory of proper scoring rules, we design a fami...

VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models

Energy-based models (EBMs) have recently been successful in representing...

Learning Priors for Adversarial Autoencoders

Most deep latent factor models choose simple priors for simplicity, trac...

Learning Hierarchical Priors in VAEs

We propose to learn a hierarchical prior in the context of variational a...

Please sign up or login with your details

Forgot password? Click here to reset