CSE3320

Machine learning

May not be offered in 1999

6 points - Two 1-hour lectures per week - Second semester - Clayton - Corequisites: CSE2309 or CSE3309 or CSC2091 or CSC3091

Objectives On completion of the subject, students should understand the main current techniques used to implement machine learning; be able to design and implement a machine learning program; and appreciate the role of machine learning in artificial intelligence theory and applications.

Synopsis This subject examines the main contending approaches to the computational modelling of learning, the applications to which they are suited and their growing use in 'mining' the data in very large corporate and governmental databases. Topics include: the need for extracting patterns from databases to support business decision making; the nature of learning and its computational modelling; symbolic approaches to machine learning; scientific discovery; information-theoretic classification; Bayesian learning; minimum encoding methods; evolutionary methods and artificial life; neural networks.

Assessment Examination (3 hours): 50% - Programming project: 50%

Recommended texts

Carbonell J G (ed.) Machine learning: Paradigms and methods MIT, 1989
Mitchell T Machine learning McGraw-Hill, 1997
Shavlik J W and Dietterich T G (eds) Readings in machine learning Morgan Kaufmann, 1990
Shrager J and Langley P (eds) Computational models of scientific discovery and theory formation Morgan Kaufmann, 1990

Back to the 1999 Science Handbook