Bridging the Gap Between f-GANs and Wasserstein GANs

by   Jiaming Song, et al.

Generative adversarial networks (GANs) have enjoyed much success in learning high-dimensional distributions. Learning objectives approximately minimize an f-divergence (f-GANs) or an integral probability metric (Wasserstein GANs) between the model and the data distribution using a discriminator. Wasserstein GANs enjoy superior empirical performance, but in f-GANs the discriminator can be interpreted as a density ratio estimator which is necessary in some GAN applications. In this paper, we bridge the gap between f-GANs and Wasserstein GANs (WGANs). First, we list two constraints over variational f-divergence estimation objectives that preserves the optimal solution. Next, we minimize over a Lagrangian relaxation of the constrained objective, and show that it generalizes critic objectives of both f-GAN and WGAN. Based on this generalization, we propose a novel practical objective, named KL-Wasserstein GAN (KL-WGAN). We demonstrate empirical success of KL-WGAN on synthetic datasets and real-world image generation benchmarks, and achieve state-of-the-art FID scores on CIFAR10 image generation.


page 7

page 15


Discriminator Contrastive Divergence: Semi-Amortized Generative Modeling by Exploring Energy of the Discriminator

Generative Adversarial Networks (GANs) have shown great promise in model...

Approximability of Discriminators Implies Diversity in GANs

While Generative Adversarial Networks (GANs) have empirically produced i...

Non-saturating GAN training as divergence minimization

Non-saturating generative adversarial network (GAN) training is widely u...

Relaxed Wasserstein with Applications to GANs

We propose a novel class of statistical divergences called Relaxed Wasse...

First Order Generative Adversarial Networks

GANs excel at learning high dimensional distributions, but they can upda...

AWGAN: Empowering High-Dimensional Discriminator Output for Generative Adversarial Networks

Empirically multidimensional discriminator (critic) output can be advant...

On Relativistic f-Divergences

This paper provides a more rigorous look at Relativistic Generative Adve...

Code Repositories


Code for "Bridging the Gap between f-GANs and Wasserstein GANs", ICML 2020

view repo


Input talk "Generative Adversarial Nets" by Marcel Neunhoeffer (University of Mannheim)

view repo

Please sign up or login with your details

Forgot password? Click here to reset