Reparameterized Sampling for Generative Adversarial Networks
This work addresses a practical bottleneck in GAN sampling for researchers and practitioners, offering an incremental improvement over existing methods.
The paper tackles the problem of poor sample efficiency in Generative Adversarial Networks (GANs) by proposing REP-GAN, a novel sampling method that uses reparameterized Markov chains in the latent space, resulting in improved sample efficiency and quality as demonstrated in experiments.
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.