3 AI mastery: Essential techniques, Part 2
This chapter covers
- An introduction to data mining
- An overview of the artificial neural networks
- A description of deep learning
- An introduction to Bayesian networks
- An overview of unsupervised learning
AI expert Arthur Samuel, introduced in chapter 1 for the success of his 1959 checkers program, defined machine learning as the field of study that gives computers the ability to learn without being explicitly programmed. “Without being explicitly programmed” can be misleading, as learning is achieved with techniques such as data mining and neural networks, which rely on algorithms explicitly programmed by engineers.
In this chapter, we will explore data mining, a technique used to extract valuable information, patterns, and associations from data. I briefly mention Bayesian networks, a method that encodes probabilistic relationships between variables of interest. I then introduce artificial neural networks and deep learning, powerful pattern recognition algorithms that have achieved impressive results in computer vision, natural language, and audio processing. We finish this chapter with unsupervised learning, a set of algorithms that can analyze unlabeled datasets to discover similarities and differences. I’ll provide enough detail to allow you to understand what these machine learning techniques entail and how they’re applied, but we won’t get caught up in the theory.