Introduction to AI

Artificial Intelligence (AI) is the branch of computer science concerned with building machines that can perform tasks which, when done by humans, are said to require intelligence — reasoning, learning, planning, perception, and language understanding.

Definitions of AI

There is no single agreed definition of AI. The classic textbook view (Russell & Norvig) groups definitions along two axes — thought vs behaviour, and human-like vs ideal/rational — giving four categories:

Modern AI mostly follows the acting rationally view — an intelligent agent perceives its environment and takes actions that maximise its chance of success.

A Simple Working Definition

"AI is the study of how to make computers do things which, at the moment, people do better." — Elaine Rich

The Turing Test (preview)

Alan Turing (1950) proposed a practical test: if a human interrogator, chatting via text with both a human and a machine, cannot reliably tell which is which, the machine is said to exhibit intelligent behaviour. (Full discussion in the "Criteria for Success" lesson.)

Goals of AI

• Build systems that can reason and draw conclusions
• Build systems that can learn from experience
• Build systems that can perceive their environment (vision, speech)
• Build systems that can understand and generate language
• Build systems that can plan and act to achieve goals

A Brief Timeline

AI is not one technique but an umbrella over search, logic, probability, learning and representation — the rest of this unit and course builds each of these pieces one at a time.