Reparameterized Sampling for Generative Adversarial Networks

TL;DR

This paper introduces REP-GAN, a reparameterized sampling method for GANs that improves sample efficiency and quality by enabling dependent proposals through Markov chain reparameterization in the latent space.

Contribution

It proposes a novel reparameterization technique for sampling in GANs, allowing dependent proposals and a closed-form acceptance ratio, enhancing sample efficiency and quality.

Findings

Significant improvement in sample efficiency over traditional methods.

Better sample quality achieved on synthetic and real datasets.

Theoretical proof of the closed-form Metropolis-Hastings ratio.

Abstract

Recently, sampling methods have been successfully applied to enhance the sample quality of Generative Adversarial Networks (GANs). However, in practice, they typically have poor sample efficiency because of the independent proposal sampling from the generator. In this work, we propose REP-GAN, a novel sampling method that allows general dependent proposals by REParameterizing the Markov chains into the latent space of the generator. Theoretically, we show that our reparameterized proposal admits a closed-form Metropolis-Hastings acceptance ratio. Empirically, extensive experiments on synthetic and real datasets demonstrate that our REP-GAN largely improves the sample efficiency and obtains better sample quality simultaneously.