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