Reparameterized Sampling for Generative Adversarial Networks

by   Yifei Wang, et al.

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.


page 14

page 17


EvolGAN: Evolutionary Generative Adversarial Networks

We propose to use a quality estimator and evolutionary methods to search...

Time Series (re)sampling using Generative Adversarial Networks

We propose a novel bootstrap procedure for dependent data based on Gener...

Improving Generative Adversarial Networks with Local Coordinate Coding

Generative adversarial networks (GANs) have shown remarkable success in ...

Exploring and Exploiting Hubness Priors for High-Quality GAN Latent Sampling

Despite the extensive studies on Generative Adversarial Networks (GANs),...

Self-Adversarially Learned Bayesian Sampling

Scalable Bayesian sampling is playing an important role in modern machin...

Sampling-Decomposable Generative Adversarial Recommender

Recommendation techniques are important approaches for alleviating infor...

Invert and Defend: Model-based Approximate Inversion of Generative Adversarial Networks for Secure Inference

Inferring the latent variable generating a given test sample is a challe...