 |
    |
Populating Genomes in a Dynamic Grid
Han Yu, Ning Jiang, and Annie S. Wu
School of Computer Science, University of Central Florida, Orlando, FL 32816-2362hyu@cs.ucf.edu
njiang@cs.ucf.edu
aswu@cs.ucf.edu
Abstract. We study the behavior of a spatially distributed GA (Collins and Jefferson 1991) that has several unique features. First, individuals are allowed to move within a two dimensional grid over time. Such movement results in a continual change in the local environment in which each individual interacts. Second, selection is directly based on an individual’s energy value and indirectly based on fitness. The fitness of an individual determines its initial energy level, but an individual’s energy level at any point during a run reflects a combination of its fitness and its interactions in the environment. Third, the population size can change dynamically throughout a run.
We study the mating radius of the individuals, and the energy-based selection which results in noisy selection. We investigate the impact of these two aspects on the selection pressure among individuals.
LNCS 3103, p. 418 f.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004