CSA/IE: Novel Clonal Selection Algorithm with Information Exchange for High Dimensional Global Optimization Problems

Zixing Cai,Xingbao Liu,Xiaoping Ren

DOI: https://doi.org/10.1007/978-3-642-33757-4_17

2012-01-01

Abstract:In order to increase the diversity of immune algorithm when solving high dimensional global optimization problems, a novel clonal selection algorithm with information exchange (CSA/IE) is proposed. The main characteristics of CSA/IE are clonal expansion and a novel hypermutation strategy. In addition, a simplex crossover operator is introduced to improve the ability of information exchange. Particularly, a novel performance evaluation criterion is constructed in this paper, by which the performance of different population-based algorithms can be compared easily.The experimental results indicate that CSA/IE outperforms that of the conventional clonal selection algorithms and the three DE variants, in terms of the performance evaluation criterion proposed. Finally, the proposed CSA/IE is generalized to optimize some hyper-high dimensional (such as 100~1000 dimensions) unimodal and multimodal test functions, and the results show that the proposed algorithm performs well in terms of the stability and the solution quality.