From GAN to WGAN
arXiv:1904.08994v1133 citations
AI Analysis
This addresses a critical bottleneck in generative modeling for researchers and practitioners, offering a more reliable method for training GANs.
The paper tackles the training instability of generative adversarial networks (GANs) by introducing Wasserstein GAN (WGAN), which uses a smooth Wasserstein distance metric to measure distribution differences, resulting in improved training stability and convergence.
This paper explains the math behind a generative adversarial network (GAN) model and why it is hard to be trained. Wasserstein GAN is intended to improve GANs' training by adopting a smooth metric for measuring the distance between two probability distributions.