Deep learning architectures have proven to be highly effective, but they are still not the best approach for one-shot learning. Different Bayesian approaches, such as the Bayesian programming language, can still beat humans at one-shot learning. In this section, we will learn about Bayesian methods and discuss the recent advancements that have been made in this domain. Additionally, we will compare the Bayesian method to a well-known technique—transfer learning—that exists in the deep learning circle to solve any problem. We will also learn when to use the one-shot approach over transfer learning.
This section comprises the following chapters:
- Chapter 5, Generative Modeling-Based Methods
- Chapter 6, Conclusions and Other Approaches
