COMP10412: Fundamentals of Artificial Intelligence (2009-2010)
The course introduces the study of Artificial Intelligence (AI) for students in all course streams. It is designed to stand alone as an introduction to AI, but also to provide a background for more advanced study. The course presents AI from a probabilistic viewpoint, and is centred around two specific problems: (i) robot localization; (ii) speech understanding. The lectures will present the main theoretical ideas needed to tackle these problems; the examples classes will re-inforce these through paper-and-pencil exercises, and the labs will involve the development of programs to solve them. There will be one hour of lectures and one hour of examples classes each week, as well as five two-hour lab sessions over the semester.
A student completing this course should:
Understand in broad outline the principal challenges of AI, the major research areas and the overall historical development of the subject.
Understand the fundamentals of probability theory, considered as an account of reasoning under uncertainty, and its central role in AI.
Understand in detail the problem of robot localization, and the methods used to solve it.
Understand in detail the problem of speech recognition, and the methods used to solve it.
Develop practical expertise in implementing probabilistic algorithms in AI.
Understand the philosophical foundations and limitations of probability theory in AI.
Assessment of Learning outcomesLearning outcomes 1-4 and 6 will be assessed by examination. Learning outcomes 5 will be assessed in the laboratory.
Contribution to Programme Learning OutcomesA1, A2, A3, A4, A5, B1, B3, C5, D4, D5, D6
Overview of Artificial Intelligence (1)
Probability and the problem of robot localization (2)
Foundations and limitations of probabilistic reasoning (1)
Introduction to speech processing and recognition (1)
Feature extraction for speech and building a simple feature-based classifier (1)
Introduction to Hidden Markov models (1)
Inference, learning and classification with hidden Markov models (2)
The role of probabilistic and non-probabilistic reasoning in other AI applications (1)
Examples classes will mirror the syllabus Laboratory exercises:
-1.1 Robot localization
-1.2 Robot localization
-2.1 Speech processing
-2.2 Speech processing
-2.3D Speech processing
Course notes will be distributed in the lectures and made available online. There is no recommended text book.