What are the two key properties of GANs that the Wasserstein GAN (WGAN) aims to improve according to the 2017 paper by Arjovsky et al.?

The Wasserstein GAN (WGAN) aims to improve the stability of the optimization process and provide a meaningful loss metric that correlates with the generator's convergence and sample quality.

The Wasserstein GAN (WGAN), introduced by Arjovsky et al. in 2017, addresses two key properties of GANs. First, it provides a meaningful loss metric that correlates with the generator's convergence and the quality of the samples it produces. Second, it improves the stability of the optimization process during GAN training. These improvements are achieved by using the Wasserstein loss function, which replaces the binary cross-entropy loss traditionally used in GANs, leading to more stable convergence.