Disentangled Representation GANs

As we've explored, GANs can generate meaningful outputs by learning the data distribution. However, there was no control over the attributes of the generated outputs. Some variations of GANs, like conditional GAN (CGAN) and auxiliary classifier GAN (ACGAN), as discussed in the previous two chapters, are able to train a generator that is conditioned to synthesize specific outputs. For example, both CGAN and ACGAN can induce the generator to produce a specific MNIST digit. This is achieved by using both a 100-dim noise code and the corresponding one-hot label as inputs. However, other than the one-hot label, we have no other ways to control the properties of generated outputs.

For a review of CGAN and ACGAN, please refer to Chapter 4, Generative Adversarial Networks (GANs), and Chapter 5, Improved GANs.

In this chapter, we will be covering the variations of GANs that enable us to modify the generator outputs. In the...