

Tutorial
Professor Riccardo Poli is a professor in the Department of Computer Science of the University of Essex where he founded the Natural and Evolutionary Computation (NEC) group. Previously at the School of Computer Science of the University of Birmingham, he has published more than 100 papers on neural networks, evolutionary algorithms (particularly genetic programming), and image/signal processing and has coauthored Foundations of Genetic Programming with Bill Langdon (Springer, February 2002). He is a cochair of EvoGP, the EvoNet working group on genetic programming, and is the chair of the GP deme at the Genetic and Evolutionary Computation Conference (GECCO) 2002 as well as programme cochair of the 2002 Foundations of Genetic Algorithms Workshop (FOGA7).
"The ABC of GP": tutorial outline
Evolutionary Computation in a Nutshell, Simple Generational GA, What is GP?
Representation, Representing Programs in Prefix Notation, Implementation, Primitive Sets, Properties of the Primitive Set, Protected Functions, Examples
GP Search Space, Examples
Initial Population in GP, Initialisation Methods, Implementation
Crossover in GP, Implementation
Mutation in GP
Running Programs in GP, Execution Strategies, Recursive Interpretation, Implementation, Examples, GP Macros
Fitness functions in GP, Symbolic Regression, Steps to Solve Symbolic Regression Problems, Examples of Symbolic Regression
Factors Determining GP Performance
Divideandconquer Strategy, Automatic Function Definition, Changes required by ADFs, example
GP Schema Theory
Conclusions
The tutorial slides can be previewed at the following URL:
http://cswww.essex.ac.uk/staff/poli/eurogp2002/
