On Stabilizing Generative Adversarial Training with Noise

Jenni, Simon; Favaro, Paolo (June 2019). On Stabilizing Generative Adversarial Training with Noise. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE Computer Society

Jenni_On_Stabilizing_Generative_Adversarial_Training_With_Noise_CVPR_2019_paper.pdf - Accepted Version
Available under License Publisher holds Copyright.
This CVPR paper is the Open Access version, provided by the Computer Vision Foundation. Except for this watermark, it is identical to the accepted version; the final published version of the proceedings is available on IEEE Xplore.

Download (1MB) | Preview

We present a novel method and analysis to train generative adversarial networks (GAN) in a stable manner. As shown in recent analysis, training is often undermined by the probability distribution of the data being zero on neighborhoods of the data space. We notice that the distributions of real and generated data should match even when they undergo the same filtering. Therefore, to address the limited support problem we propose to train GANs by using different filtered versions of the real and generated data distributions. In this way, filtering does not prevent the exact matching of the data distribution, while helping training by extending the support of both distributions. As filtering we consider adding samples from an arbitrary distribution to the data, which corresponds to a convolution of the data distribution with the arbitrary one. We also propose to learn the generation of these samples so as to challenge the discriminator in the adversarial training. We show that our approach results in a stable and well-behaved training of even the original minimax GAN formulation. Moreover, our technique can be incorporated in most modern GAN formulations and leads to a consistent improvement on several common datasets.

Item Type:

Conference or Workshop Item (Paper)


08 Faculty of Science > Institute of Computer Science (INF)

UniBE Contributor:

Jenni, Simon and Favaro, Paolo


000 Computer science, knowledge & systems
500 Science > 510 Mathematics


IEEE Computer Society




Xiaochen Wang

Date Deposited:

18 Feb 2020 09:20

Last Modified:

18 Feb 2020 09:20





Actions (login required)

Edit item Edit item
Provide Feedback