Transfer learning for GANs


#1

Are there any examples of people applying transfer learning to GANs?

We’ve seen that using transfer learning for classifiers allows you to build a good classifier even if you don’t have a lot of data. It’d be really interesting to be able to do something similar for generation, but it’s not obvious to me how to go about this. Do you replace the top layers of both the discriminator and generator? Leave all the layers trainable after that? etc.