A New GEP Algorithm Based on Multi-phenotype Chromosomes

This paper presents an improved gene expression programming (GEP) algorithm based on multi-phenotype chromosomes (MPC-GEP). The populations in MPC-GEP are composed of chromosomes with multiple phenotypes. Each multi-gene chromosome corresponds to multiple expression trees. The new algorithm can find the optimal individual in less time than traditional GEP. Finally, experiments on the new algorithm against traditional GEP algorithm are conducted on several benchmark datasets. Results show that MPC-GEP outperforms traditional GEP in function finding in terms of speed.