Generative algorithms are part of unsupervised learning techniques. They underpin one of the most innovative concepts in machine learning in the past decade: Generative Adversarial Networks (GANs). In this chapter, we will be looking at the variations and developments in generative models in recent times.
A generative model can learn to mimic any distribution of it. Their potential is huge as they can be taught to recreate similar models in any domain. Some of these domains include, but are not limited, to the following:
- Images
- Music
- Speech
- Text
- Videos
There are a host of published papers outlining the advancements in GANs, and links to some of those most noteworthy have been listed at the end of this chapter.
Specifically, we will be covering the following topics in this chapter:
- Discriminative versus generative algorithms
- Different types...
