GAN variations and timelines

There have been many significant developments to GAN research in recent times. The following timeline shows some of the most noteworthy advances:

This chapter will now give insight into these developments, their applications, and results.

Conditional GANs

Conditional GANs are a central theme that form the building blocks of many state-of-the-art GANs. The paper submitted by Mirza and Osindero in 2014 shows how integrating the class labels of data yields greater stability in GAN training. This idea of conditioning GANs with prior information is a common approach in future GAN research. It is particularly important for papers whose main focus is on image-to-image or text-to-image applications:

...