An improvement of opposition-based differential evolution with archive solutions
Jun-ichi Kushida, Akira Hara, Tetsuyuki Takahama · 2014
Differential evolution (DE) is a simple yet efficient evolutionary algorithm. Because of its simplicity, effectiveness and robustness, DE has gradually become more popular and applied in various fields. In addition, a lot of works have been done to improve the search ability of DE. Among them, opposition-based DE (ODE), which is incorporated opposition-based learning (OBL), has shown better performance compared to classical DE. The main idea behind OBL is the simultaneous consideration of an estimate and its corresponding opposite estimate in order to achieve a better approximation for the current candidate solution. In this paper, we improve OBL by using archive solutions and propose an improved version of the ODE. Experimental verifications are conducted on well-known benchmark functions and the performance of the proposed method is evaluated by comparing with classical DE and generalized ODE.