GAN on Atari images

Almost every book about DL uses the MNIST dataset to show you the power of DL, which, over the years, has made this dataset extremely boring, like a fruit fly for genetic researchers. To break this tradition, and add a bit more fun to the book, I’ve tried to avoid well-beaten paths and illustrate PyTorch using something different. I briefly referred to generative adversarial networks (GANs) earlier in the chapter. In this example, we will train a GAN to generate screenshots of various Atari games.

The simplest GAN architecture is this: we have two NNs where the first works as a ”cheater” (it is also called the generator), and the other as a ”detective” (another name is the discriminator). Both networks compete with each other — the generator tries to generate fake data, which will be hard for the discriminator to distinguish from your dataset, and the discriminator tries to detect the generated data samples. Over...