Fetching the paper…

Gradient Layer: Enhancing the Convergence of Adversarial Training for Generative Models · Around