Implicit competitive regularization in GANs
To improve the stability of GAN training we need to understand why they can produce realistic samples. Presently, this is attributed to properties of the divergence obtained under an optimal discriminator. This argument has a fundamental flaw: If we do not impose regularity of the discriminator, it can exploit visually imperceptible errors of the generator to always achieve the maximal generator loss. In practice, gradient penalties are used to regularize the discriminator. However, this needs a metric on the space of images that captures visual similarity. Such a metric is not known, which explains the limited success of gradient penalties in stabilizing GANs. We argue that the performance of GANs is instead due to the implicit competitive regularization (ICR) arising from the simultaneous optimization of generator and discriminator. ICR promotes solutions that look real to the discriminator and thus leverages its inductive biases to generate realistic images. We show that opponent-aware modelling of generator and discriminator, as present in competitive gradient descent (CGD), can significantly strengthen ICR and thus stabilize GAN training without explicit regularization. In our experiments, we use an existing implementation of WGAN-GP and show that by training it with CGD we can improve the inception score (IS) on CIFAR10 for a wide range of scenarios, without any hyperparameter tuning. The highest IS is obtained by combining CGD with the WGAN-loss, without any explicit regularization.
Code (3)
Tasks
Image GenerationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Adversarial Lipschitz Regularization
Generative adversarial networks (GANs) are one of the most popular approaches when it comes to training generative models, among which variants of Wasserstein GANs are considered superior to the standard GAN formulation …
Image GenerationViTGAN: Training GANs with Vision Transformers
Recently, Vision Transformers (ViTs) have shown competitive performance on image recognition while requiring less vision-specific inductive biases. In this paper, we investigate if such performance can be extended to ima…
Image GenerationHow Well Generative Adversarial Networks Learn Distributions
This paper studies the rates of convergence for learning distributions implicitly with the adversarial framework and Generative Adversarial Networks (GANs), which subsume Wasserstein, Sobolev, MMD GAN, and Generalized/Si…
Density EstimationvalidDISTRIBUTIONAL CONCAVITY REGULARIZATION FOR GANS
We propose Distributional Concavity (DC) regularization for Generative Adversarial Networks (GANs), a functional gradient-based method that promotes the entropy of the generator distribution and works against mode collap…
Kernel-Guided Training of Implicit Generative Models with Stability Guarantees
Modern implicit generative models such as generative adversarial networks (GANs) are generally known to suffer from issues such as instability, uninterpretability, and difficulty in assessing their performance. If we see…