Enhancing Differential Evolution with Commensal Learning and Uniform Local Search

Abstract

Differential evolution (DE) is a popular and powerful evolutionary algorithm for global optimization problems. However, the combination of mutation strategies and parameter settings of DE is problem dependent and choosing the suitable one is a challenge work and timeconsuming. In addition, the deficiency in local exploitation also has a significant influence on the performance of DE. In order to solve these problems, a DE variant with Commensal learning and uniform local search (CUDE) has been proposed in this paper. In CUDE, commensal learning is proposed to adaptively select optimal mutation strategy and parameter setting simultaneously under the same criteria. Moreover, uniform local search enhances exploitation ability. Comprehensive experiment results on all the CEC 2013 test suite and comparison with the state-of-the-art DE variants indicate that the CUDE is very competitive.