Overview: Improving Our First GAN

In this chapter, we will learn about the main challenges in training and understanding GANs, as well as how to solve them. We will learn about vanishing gradients, mode collapse, training instability, and other challenges. We will also learn about multiple deep-learning model architectures that have been successful using the GAN framework. Furthermore, we will learn to possibly improve our first GAN by implementing new loss functions and algorithms.

Topics covered in this chapter

In this chapter, we will continue to focus on the CIFAR-10 dataset and cover the following topics:

• Challenges in training GANs

• Tricks of the trade

• GAN model architectures

• GAN algorithms and loss functions