Fuzzy-Differential Coevolution for Many Objective Optimization
Selina Khoirom, Pratyusha Rakshit · 2024
The paper proposes a novel coevolution-based many- objective optimization (MaOO) algorithm to enjoy the implicit parallelism of the evolutionary algorithm. The proposed MaOO handles individual objectives in parallel using differential evolution (DE) algorithm. After achieving convergence, a group of good-quality solutions is judiciously selected from each DE population, dealing with a specific objective. Next, the groups are combined to form a union set. Finally, a fuzzy membership- induced rank measure is devised to identify the best-ranked equally good members from the union set indicating the approximate Pareto optimal solutions to the given MaOO problem. The proposed MaOO algorithm, referred to as fuzzy- differential coevolution (FDC) is compared with three state-of-the-art techniques. Experiments undertaken reveal that FDC outperforms its contenders with respect to the traditional performance metrics.