GAN improvements that lead towards finding global minima for any given distribution mapping. GANs are versatile but still suffer from mode collapse and instabi

GAN improvements that lead towards finding global minima for any given distribution mapping. GANs are versatile but still suffer from mode collapse and instability, drastically reducing their utility.