Timezone: »

 
Poster
Understanding Noise Injection in GANs
Ruili Feng · Deli Zhao · Zheng-Jun Zha

Wed Jul 21 09:00 PM -- 11:00 PM (PDT) @
in Poster Session 4 »

Noise injection is an effective way of circumventing overfitting and enhancing generalization in machine learning, the rationale of which has been validated in deep learning as well. Recently, noise injection exhibits surprising effectiveness when generating high-fidelity images in Generative Adversarial Networks (GANs) (e.g. StyleGAN). Despite its successful applications in GANs, the mechanism of its validity is still unclear. In this paper, we propose a geometric framework to theoretically analyze the role of noise injection in GANs. First, we point out the existence of the adversarial dimension trap inherent in GANs, which leads to the difficulty of learning a proper generator. Second, we successfully model the noise injection framework with exponential maps based on Riemannian geometry. Guided by our theories, we propose a general geometric realization for noise injection. Under our novel framework, the simple noise injection used in StyleGAN reduces to the Euclidean case. The goal of our work is to make theoretical steps towards understanding the underlying mechanism of state-of-the-art GAN algorithms. Experiments on image generation and GAN inversion validate our theory in practice.

Keywords: Deep Learning Theory

Author Information

Ruili Feng (USTC)
Deli Zhao (Alibaba Group)
Zheng-Jun Zha (University of Science and Technology of China)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors