Computational Intelligence

Razvan Andonie

Angel Cataron

Overview

The course introduces the foundations of artificial neural networks, genetic algorithms, fuzzy systems, swarm intelligence, and hybridizations of the above techniques. This domain is called Computational Intelligence, and it is a numerical interpretation of biological intelligence.

Contents

*      Preliminaries (2 hours)

*      Basic Concepts (6 hours)

*      Single-Layer Perceptrons as Classifiers (5 hours)

*      Multi-Layer Feedforward Networks (6 hours)

*      Single-Layer Feedback Networks (6 hours)

*      Associative Memories (5 hours)

*      Self-Organizing Networks (6 hours)

*      Genetic Algorithms (3 hours)

*      Swarm Intelligence (2 hours)

*      Fuzzy Logic (3 hours)

*      Fuzzy Neural Networks (2 hours)

*      Radial Basis Function Neural Networks (2 hours)

*      The Power and Computational Complexity of Neural Networks (2 hours)

*      Epistemological Considerations of Neural Computing (2 hours)

Prerequisites

*      Algorithms and Complexity

*      Programming

Textbooks

We are going to use results presented in several textbooks and journal papers. Just to make your life easy, here are the recommended textbooks:

*      Andonie, R., Cataron, A. Computational Intelligence (in Romanian), Transylvania University Press, Brasov, Romania, 2002.

*      Zurada, J. Introduction to Artificial Neural Systems. West Publishing Company, St. Paul, 1992. I have used this textbook for several years in Brasov and I still consider it the most appropriate for our department here.

*      Haykin, S. Neural Networks - A Comprehensive Foundation. Macmillan College Publishing Company, New York, 1999 (second edition). It is generally considered as one of the best.

*      M. T. Hagan, H. B. Demuth, M. Beale, Neural Network Design, PWS Publishing Co., Boston, 1996. This is the textbook I have used in San Antonio, Texas.

*      Rao, V. C++ Neural Networks and Fuzzy Logic, M&T Books, IDG Books Worldwide, Inc., 1995. This book can be used as a handbook for neural OOP implementations. It can be downloaded from http://freebooks.by.ru/

Links

*      DSI Neural Networks Group is a very complete source of electronic information

*      Gurney's Notes help you to learn about the basic concepts and ideas in artificial neural networks

*      IEEE Neural Networks Council

*      Software for Genetic Algorithms

*      Stuttgart Neural Network Simulator

*      Neural Networks Commercial Software Tools

*      Sullivan's Primer on Genetic Algorithms

*      Machine Learning Datasets is a very good source of benchmarks for testing your neural implementations.

*      Ant Colony Optimization is a source of tutorials.