Evolving problems to learn about particle swarm and other optimisers

Author

Langdon, W.B. ; Poli, Riccardo

Author_Institution

Dept. of Comput. Sci., Essex Univ., Colchester

Volume

1

fYear

2005

fDate

5-5 Sept. 2005

Firstpage

81

Abstract

We use evolutionary computation (EC) to automatically find problems which demonstrate the strength and weaknesses of modern search heuristics. In particular we analyse particle swarm optimization (PSO) and differential evolution (DE). Both evolutionary algorithms are contrasted with a robust deterministic gradient based searcher (based on Newton-Raphson). The fitness landscapes made by genetic programming (GP) are used to illustrate difficulties in GAs and PSOs thereby explaining how they work and allowing us to devise better extended particle swarm systems (XPS)

Keywords

Newton-Raphson method; genetic algorithms; gradient methods; particle swarm optimisation; search problems; Newton-Raphson-based method; differential evolution; evolutionary algorithm; evolutionary computation; extended particle swarm system; genetic programming; particle swarm optimisation; robust deterministic gradient based searcher; search heuristics; Computer science; Evolutionary computation; Genetic mutations; Genetic programming; Mathematical analysis; Optimization methods; Particle swarm optimization; Robustness; Stability; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolutionary Computation, 2005. The 2005 IEEE Congress on

Conference_Location

Edinburgh, Scotland

Print_ISBN

0-7803-9363-5

Type

conf

DOI

10.1109/CEC.2005.1554670

Filename

1554670