Optimal Population Size and the Genetic Algorithm

We conduct experiments to determine the optimum population size for problems as the instance size varies. We show that increasing the population size increases the accuracy of the GA. Increasing population size also causes the number of generations to converge to increase. The optimal population for a given problem is the point of inflection where the benefit of quick convergence is offset by increasing inaccuracy. Finally, we propose a method that might be used for determining the optimum population size for a given problem instance. This method holds for all three of the dissimilar problems that were used to conduct the experiment. It seems possible that it may hold for all GA applicable problems. Key-Words: Genetic Algorithm, Population, Optimization, Evolutionary Computation