Zusammenfassung für den Vortrag am 15.09.2022 (17:00 Uhr)Math Machine Learning seminar MPI MIS + UCLA
Lisa Maria Kreusser (University of Bath)
Wasserstein GANs Work Because They Fail (to Approximate the Wasserstein Distance)
Siehe auch die Vortragsfolien dieses Vortrages.
Wasserstein GANs (WGANs) are based on the idea of minimising the Wasserstein distance between a real and a generated distribution. In this talk, we provide an in-depth mathematical analysis of differences between the theoretical setup and the reality of training WGANs. We gather both theoretical and empirical evidence that the WGAN loss is not a meaningful approximation of the Wasserstein distance. Moreover, we argue that the Wasserstein distance is not even a desirable loss function for deep generative models, and conclude that the success of WGANs can be attributed to a failure to approximate the Wasserstein distance.