CS6759 Course Outline
Artificial Neural Networks
COURSE DESCRIPTION
This course introduces and relates the basic concept of
neural networks. The course will provide a current and coherent view
of artificial neural networks. The neural net
algorithms will be discussed to understand contemporary neurocomputing
technology.
It emphasizes mathematical analysis of neural networks,
methods for training networks, and application of networks to practical
problems. Neural network implementation will be discussed to understand
contemporary neurocomputing and softcomputing techniques.
Students in computer science, engineering, and behavioral science can obtain
immediate benefits to visualize new approaches and interdisciplinary
perspectives.
Course Note: see /usr/local/pub/cs/cs6759
REFERENCE BOOK:

J. Freeman and D. Skapura, ``Neural Networks: Algorithm,
Applications, and Programming Techniques'',
AddisonWesley, 1992,

YH Pao,
``Adaptive Pattern Recognition and Neural Networks''
AddisonWesley, 1989,

R. Schalkoff ``Pattern Recognition: Statistical Structural and Neural
Approaches'', John Wiley, 1992,

M. Hagan, H. Demuth, and M. Beale
``Neural Network Design'', PWS, 1996.

L. Fausett ``Fundamentals of Neural Networks: Architectures, Algorithm,
and Applications'',
Prentice Hall, 1994.

J. Principe, N. Euliano, and C. Lefebvre ``Neural and Adaptive Systems''
John Wiley and Sons, Inc, 1999.
GRADING SCHEMA
Lab/Assignment 30%
Research Survey and Presentation 10%
Term Project /Presentation 30%
Final Exam 30%

4 labs & assignments.

Must complete term project & presentation
TIME TABLE

a) Course information
b) Introduction to Neuro c) perceptron

Generalized Delta Rule

Associate Memory
a)Linear Associators (BAM) b) Matrix Associative Memory

Hopfield Network
a) Discrete Hopfield Networks,
b) Continuous Hopfield Networks, c) TravelingSalesperson Problem

Instar, outstar, Learning Vector Quantalization (LVQ, LVQ2, LVQ3)

SelfOrganize Map(Kohonon SOM)

Adaptive Resonance Theory (ART1, ART2)

Recurrent Neural networks

Neurocognitron

Speach recognition by Spatiotemporal Neural networks

a) Boltzmann Machine, b) Counterpropagation Networks