Dynamic perturbation for population diversity management in differential evolution
Le Van Cuong, Nguyen Ngoc Bao, Nguyen Khanh Phuong, Huỳnh Thị Thanh Bình · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2022
The performance of Differential Evolution (DE) is closely related to the population diversity since its mechanism of generating offspring depends wholly on the differences between individuals. This paper presents a simple perturbation technique to maintain the population diversity in which the noise intensity is adjusted dynamically during the search. A modification of the well-known L-SHADE adaptation method is also introduced to manipulate the convergence behaviour of DE. By incorporating these techniques, we develop a new variant of DE called S-LSHADE-DP. Experiment results conducted on the benchmark suite of CEC '22 competition show that S-LSHADE-DP is highly competitive with current state-of-the-art DE-based algorithms. The implementation of S-LSHADE-DP is available at https://github.com/cuonglvsoict/S-LSHADE-DP.