Chapter 4 - Machine Learning

Abstract

This chapter gives an account of machine learning, which is the subfield of artificial intelligence (AI) concerned with intelligent systems that can learn. This chapter adopts a view of intelligent systems as agents. Learning is often viewed as the most fundamental aspect of intelligence because it enables the agent to become independent of its creator. It is an essential component of an agent design whenever the designer has incomplete knowledge of the task environment. Therefore, learning provides autonomy in that the agent is not dependent on the designer's knowledge for its success and can free itself from the assumptions built into its initial configuration. Learning in intelligent agents is essential both as a construction process and as a way to deal with unknown environments. Learning agents can be divided conceptually into a performance element, which is responsible for selecting actions, and a learning element, which is responsible for modifying the performance element. The nature of the performance element and the kind of feedback available from the environment determine the form of the learning algorithm. The chapter develops a comprehensive theory of the complexity of induction, which analyzes the inherent difficulty of various kinds of learning problems in terms of sample complexity and computational complexity.