Empirical modelling of genetic algorithms

This paper addresses the problem of reliably setting genetic algorithm para meters for consistent labelling problems. Genetic algorithm parameters are notoriously difficult to determine. This paper proposes a robust empirical framework, based on the analysis of factorial experiments. The use of a gra eco-latin square permits an initial study of a wide range of parameter sett ings. This is followed by fully crossed factorial experiments with narrower ranges, which allow detailed analysis by logistic regression. The empirica l models derived can be used to determine optimal algorithm parameters and to shed light on interactions between the parameters and their relative imp ortance. Refined models are produced, which are shown to be robust under ex trapolation to up to triple the problem size.