FUZZY LOGIC, NEURAL NETWORKS AND SOFT COMPUTING
Course Control No.: 24936 2-4 units
Spring, 1996 2 hours lecture
DESCRIPTION: This course provides an introduction to soft computing
a collection of methodologies which underlie the conception, design and
deployment of intelligent systems. The principal components of soft
computing are fuzzy logic, neural network theory and probabilistic reasoning,
with the latter subsuming genetic algorithms, evidential reasoning and belief
networks. Within fuzzy logic, attention is focused on the calculus of fuzzy
if-then rules -- a basic tool which is employed extensively in a
wide variety of applications ranging from consumer electronics to
medical diagnostic systems. In combination with neural network
techniques, the calculus of fuzzy if-then rules provides a basis for
the design of neuro-fuzzy systems which have the capability to learn
and to adapt to changes in operating conditions. The basics of neural
network theory, genetic algorithms and probabilistic reasoning are discussed
and their roles in the applications of soft computing are illustrated by
practical examples.
PREREQUISITES: The course is self-contained. No prior knowledge of fuzzy
logic or neural network theory is required.
FINAL EXAMS: Term paper or programming project.
INSTRUCTOR IN CHARGE: L. A. Zadeh: 642-4959; zadeh@cs
TIME AND PLACE: Mondays 2-4 pm; 373 Soda Hall