Hands-On Generative Adversarial Networks with Keras

All about that inductive bias

In their paper Deep Image Prior, Dmitry Ulyanov et al. show that a randomly-initialized neural network can be used as a handcrafted prior with excellent results in standard inverse problems such as denoising, super-resolution, and inpainting. This evidence, and the evidence brought by the paper Are GANs created Equal?, leads us to question how important the GAN framework is with respect to the convolutional architectures that are being used to solve these problems. We should bear this in mind when developing new GAN architectures.