A Novel Multigroup Evolutionary Algorithm Based on Competition and Communication among Population
Xiangjun Wang · Acta Simulata Systematica Sinica · 2005
Evolution is not only a natural selection procession, but also the result of competition and communication among populations. A novel improved multigroup evolutionary programming algorithm is proposed based on this viewpoint. In the algorithm, the scale of population is determined by group competition, and the mutation of population is determined by the scale of population. The individual, together with information, is exchanged while the population scale is variable. The simulations based on benchmarks confirms that this algorithm is better than classic evolutionary programming algorithm in the aspects of global optimization, convergence speed and the robustness