Efficient Sampling for Generative Adversarial Networks with Reparameterized Markov Chains

Yifei Wang, Yisen Wang, Jiansheng Yang, Zhouchen Lin

Keywords: Generative Adverarial Networks, Sampling, Markov chain Monte Carlo, Reparameterization

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.

One-sentence Summary: We develop a novel sampling method for GANs, called REP-GAN, that enjoys better sample efficiency and sample quality.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=q7oPrE8Yac