Study on Modular Mechanism of the Co-Evolution Genetic Algorithm

From ecology, physiology and computable learning theory, the foundation and characters of the modular methods, which is meaningful in the co-evolution genetic algorithm (CGA) are discussed, and task decomposition in CGA is discussed. It is proven through analyses and examples that reasonable task decomposition is the key in applying CGA successfully.