Aims

• To introduce the fundamental insights, concepts and techniques of Artificial Intelligence.
• To give experience of designing, writing and debugging simple AI programs.
• To give an appreciation of the difficulties of, and different approaches to, the construction of intelligent agents.

Objectives

• Devise appropriate representations for reasoning about a problem
• Design and implement simple programs in Pop-11
• Plan a project and write an appropriate report on its outcome
• Implement an AI technique or algorithm
• Explain the trade-offs between different approaches to the same problem
• Demonstrate familiarity with the major subfields in AI, and awareness of AI applications.

Prerequisites

None. In particular no previous programming experience is assumed.

Teaching Methods

11 Lectures
22 Laboratory sessions

The lab sessions will be used to learn Pop-11 using on-line teaching material, supported by demonstrators. The project (assessed) will be carried out in these sessions during the second half of the semester. Additional personal study of about 2.5 hours a week is recommended.

Assessment

100% continuous

A 2500 word report on an AI programming project. The project will be carried out in supervised lab sessions and in the student's own time. The deadline for the hand-in of the report is at the end of Week 12.

Recommended Books

Title	Author(s)	Publisher	Comments
Lecture notes on AI and Pop-11 for SEMQAI	Jeremy Wyatt	School of Computer Science	Available to students on the course
On-line teaching materials for Pop-11	Various	School of Computer Science	Available to students on the course

1. Introduction (1 lecture)
   • A history of mind
   • Fundamental concepts in AI
   • Goals and achievements of AI

2. Pop-11 (2 lectures)
   • Data objects
   • Operations
   • The stack
   • Errors
   • Variables
   • Lists
   • Procedures

3. Reactive Agents (1 lecture)
   • Types of agents
   • Reactive Agents
   • How to build a reactive agent in Pop-11
   • The Pop-11 pattern matcher

4. Reasoning Agents (2 lectures)
   • Representing the world
   • The Pop-11 database
   • Predicates and sentences
   • Reasoning with representations
   • Search

5. Learning Agents (2 lectures)
   • Definitions of learning
   • Learning in animals
   • Simple learning machines
   • Generalisation
   • Neural networks
   • Learning in Artificial Neural Networks

6. Evolved Agents (1 lecture)
   • Natural Evolution
   • Artificial Evolution
   • Genetic Algorithms
   • Applications

7. Computer Vision (1 lecture)
   • Vision in animals
   • Low level computer vision
   • Applications of computer vision

8. Understaning Language (1 lecture)
   • Context Free grammars
   • Parse Trees