TEACHING NEURAL NETWORKS CONCEPTS AND THEIR LEARNING TECHNIQUES

Neural networks have become increasing popular in the fields of Science and Engineering over the last decade.Most graduate schools in the United States of America and probably in other parts of the world have started offering neural networks as a graduate/postgraduate course.Neural networks are used for nonlinear systems modeling, estimation and prediction of parameters, pattern matching, identification and control.It is necessary that engineering students learn the basics of neural networks somewhere in their undergraduate degree program without taking a full quarter/semester course.This paper presents a software developed in Java to teach basic neural network concepts with backpropagation based learning in a couple of weeks for undergraduate engineering students.This can be done as part of a modeling and simulation course in various disciplines of Engineering.