Introduction to Generative Models

In this chapter, you will learn the basics of generative models. We will start with a brief description of, and comparison between, discriminative and generative models, in which you will learn about the properties of these models. Then, we will focus on a comparison between generative models, and briefly describe how they have been used to achieve state-of-the-art models in fields such as computer vision and audio.

We will also cover other models, and then we will focus on the building blocks of Generative Adversarial Networks (GANs), their strengths, and limitations. This information is valuable because it can inform our decisions when approaching a machine learning problem with GANs, or when learning some new development in GANs.

We will cover the following topics as we progress with this chapter:

• Discriminative and generative models compared
• Generative models
• GANs – building blocks
• GANs – strengths and weaknesses