Wasserstein GAN

WGAN is an option to regular GAN training. WGANs have loss functions that correlate with image quality. Additionally, the stability of the training improves and is not as dependent on the architecture and provide significant learning curves useful for debugging.

For further reading on Wasserstein GAN refer to https://arxiv.org/abs/1701.07875

Get Neural Network Programming with TensorFlow now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.

Start your free trial