These four lectures give an introduction to basic (artificial) neural network architectures and learning rules.
We will show how to construct a set of simple artificial "neurons" and train them to serve a useful function. Such neural networks have been applied in the aerospace, automotive, banking, defense, electronics, entertainment, financial, insurance, manufacturing, oil and gas, robotics, telecommunications and transportation industries.
TEXT: Neural Network Design by Hagan, Demuth and Beale, ISBN 0-9717321-0-8, available from John Stovall at the CU bookstore, firstname.lastname@example.org, phone 303-492-3648
The FIRST FOUR CHAPTERS of the text are available at: http://hagan.okstate.edu/nnd.html in PDF format.
OVERHEADS for this course were created by Professor Martin Hagan, Oklahoma State University. They contain examples similar to but not the same as those in the text. It will be helpful to print your own copy of these overheads so you can follow the lectures with ease. The full set overheads are aslo available on the web at: http://hagan.okstate.edu/nnd.html. These are in both Powerpoint format, and as PDF files that are archived with Stuffit.
You can directly get the PDF versions of the slides that will be used in this course (without using Stuffit) from the below. Most of the ones used in the lectures will be from Chapters 2-4, with some from 10 - 12.