Timezone: »

 
Poster
Perceptual Generative Autoencoders
Zijun Zhang · Ruixiang ZHANG · Zongpeng Li · Yoshua Bengio · Liam Paull

Thu Jul 16 07:00 AM -- 07:45 AM & Thu Jul 16 08:00 PM -- 08:45 PM (PDT) @
in Poster Session 42 »

Modern generative models are usually designed to match target distributions directly in the data space, where the intrinsic dimension of data can be much lower than the ambient dimension. We argue that this discrepancy may contribute to the difficulties in training generative models. We therefore propose to map both the generated and target distributions to the latent space using the encoder of a standard autoencoder, and train the generator (or decoder) to match the target distribution in the latent space. Specifically, we enforce the consistency in both the data space and the latent space with theoretically justified data and latent reconstruction losses. The resulting generative model, which we call a perceptual generative autoencoder (PGA), is then trained with a maximum likelihood or variational autoencoder (VAE) objective. With maximum likelihood, PGAs generalize the idea of reversible generative models to unrestricted neural network architectures and arbitrary number of latent dimensions. When combined with VAEs, PGAs substantially improve over the baseline VAEs in terms of sample quality. Compared to other autoencoder-based generative models using simple priors, PGAs achieve state-of-the-art FID scores on CIFAR-10 and CelebA.

Keywords: Deep Generative Models · Autoencoders · Deep Learning - Generative Models and Autoencoders

Author Information

Zijun Zhang (University of Calgary)
Ruixiang ZHANG (Mila/UdeM)
Zongpeng Li (Wuhan University)
Yoshua Bengio (Montreal Institute for Learning Algorithms)
Liam Paull (Université de Montréal)

More from the Same Authors