An Adaptive Neighboring Search using Crossover-like Mutation for Function Optimization

We propos巴 anew population-based Evolutionary AIgorithm, which us巴sr巴al-じodedrepresenlation and normal distribution crossover-like mutation for gen巴ratlngnext searching points. This Gaussian dislribution is form巴d based on lh巴 positionrelationships b巴tweenan individual and ilS neighbors, and is not carri巴dwith self-adapting parameters as inheritable traits. This algorithm causes emergence of c1usters of individuals within the population, as the result of巴volutionsof each individuals without int巴nt to c1uster. Through searching indep巴ndently,that em巴rg巴nt clust巴rsintroduce various solutions that incJude optimum at the same time,巴venif the problem has strong local minlma