UNSW Handbook Course - Neural Networks

Neural Networks - COMP9444


	Faculty: Faculty of Engineering
	School: School of Computer Science and Engineering
	Contact: Wilson,William Hulme
	Campus: Kensington Campus
	Career: Postgraduate
	Units of Credit: 6
	EFTSL: 0.125 (more info)
	Contact Hours per Week: 3
	Enrolment Requirements:
	Prerequisite: COMP2011 or COMP2711 or COMP9024, and 12uc COMP3### or COMP4### or COMP9### - excluding Group A, or enrolment in MITprogram 8684 or GradCert program 7344.
	Session Offered: See Class Timetable
	Fee Band: 2 (more info)

Description

Topics chosen from: Network architectures: perceptrons, Hopfield and Kohonen nets, ART models, back-propagation trained feed-forward networks, recurrent nets, weightless nets. Computational complexity analysis of training neural network architectures. Probabilistic analysis of generalisation capabilities of feed-forward networks. Hardware based neural nets. Introduction to fuzzy logic, neurotrained nets; designing successful applications of neural networks; tensor product networks; and recent developments in neural networks. The assessment will include a lab project related to application of neural nets.

Further Information

CSE class page: www.cse.unsw.edu.au/~cs9444

URL for this page:

© The University of New South Wales (CRICOS Provider No.: 00098G), 2004-2011. The information contained in this Handbook is indicative only. While every effort is made to keep this information up-to-date, the University reserves the right to discontinue or vary arrangements, programs and courses at any time without notice and at its discretion. While the University will try to avoid or minimise any inconvenience, changes may also be made to programs, courses and staff after enrolment. The University may also set limits on the number of students in a course.