A Novel Clone Selection Algorithm for High-Dimensional Global Optimization Problems

The clone selection algorithm (CSA) is a stochastic, population-based evolutionary method that can be applied to the global optimization problems. The paper proposes a variation on the traditional CSA: clone selection algorithm with simplex crossover, or CSA_SPX. The novel algorithm employs the randomized distribution scheme for clone individuals, bit hyper-mutation and simplex crossover to significantly improve the performance of the original algorithm. Application of the CSA_SPX on 23 benchmark optimization problems shows a marked improvement in performance over the traditional CSA.