class period: Mondays, 7:00pm - 9:45pm mdt (6:00pm - 8:45pm pdt)
office hours: Mondays, 6:00pm - 7:00pm mdt (5:00pm - 6:00pm mdt) (please make an appointment)

Catalog data:
CS/ECE 578 Neural Network Design:
Introduction to neural networks and problems that can be solved by their application; introduction of basic neural network architectures; learning rules are developed for training these architectures to perform useful functions; various training techniques employing the learning rules discussed and applied; neural networks used to solve pattern recognition and control system problems. Prereq: perm. Prereq: Graduate standing or instructor permission. 

This course was offered last time in Spring 2006.

We will talk about artificial neural networks as universal approximators and classifiers. We will cover various learning (training) algorithms for different applications such as object (face, image, handwriting) recognition, guidance of aerial/underwater unmanned vehicles, robotics problems, intelligent control, in general, systems capable of learning and making intelligent decisions on their own. At the end of the course, we will also cover fuzzy systems for similar applications.

Class description (pdf)
Class policy (pdf )

Instructor:
Milos Manic, Ph.D.
University of Idaho