MR-GAN: Manifold Regularized Generative Adversarial Networks
This addresses the problem of unstable GAN training for researchers and practitioners, though it appears incremental as it builds on existing GAN frameworks.
The paper tackles the challenge of training generative adversarial networks (GANs) by proposing a manifold regularization term that forces the generator to respect the geometry of real data, leading to improved generalization, stability, and avoidance of mode collapse.
Despite the growing interest in generative adversarial networks (GANs), training GANs remains a challenging problem, both from a theoretical and a practical standpoint. To address this challenge, in this paper, we propose a novel way to exploit the unique geometry of the real data, especially the manifold information. More specifically, we design a method to regularize GAN training by adding an additional regularization term referred to as manifold regularizer. The manifold regularizer forces the generator to respect the unique geometry of the real data manifold and generate high quality data. Furthermore, we theoretically prove that the addition of this regularization term in any class of GANs including DCGAN and Wasserstein GAN leads to improved performance in terms of generalization, existence of equilibrium, and stability. Preliminary experiments show that the proposed manifold regularization helps in avoiding mode collapse and leads to stable training.