Signi cance of Locality and Selection Pressure in the Grand Deluge Evolutionary Algorithm

This paper presents the results of a parameter study of the Grand Deluge Evolutionary Algorithm, whose special features consist of local interactions between individuals within a spatially structured population and a self{adjusting control mechanism of the selection pressure. Since both ingrediences are parametrizable this study aims at the identi cation of the signi cance and sensitivity of the parameter settings with regard to the performance of the algorithm, especially under the transition from one{ to two{dimensional neighborhood patterns.