A Gauss-Newton Approach for Min-Max Optimization in Generative Adversarial Networks

arXiv:2404.07172v1 Announce Type: new
Abstract: A novel first-order method is proposed for training generative adversarial networks (GANs). It modifies the Gauss-Newton method to approximate the min-max Hessian and uses the Sherman-Morrison inversion formula to calculate the inverse. The method corresponds to a fixed-point method that ensures necessary contraction. To evaluate its effectiveness, numerical experiments are conducted on various datasets commonly used in image generation tasks, such as MNIST, Fashion MNIST, CIFAR10, FFHQ, and LSUN. Our method is capable of generating …

abstract adversarial arxiv cs.lg cs.na fixed-point gans gauss generative generative adversarial networks math.na math.oc max min networks novel optimization training type