An Opposition-Based Learning Archerfish Hunting Optimizer for Global Optimization
Aridj Ferhat, Farouq Zitouni, Abdelhadi Limane, Rihab Lakbichi, Saad Harous · 2024
This paper introduces OBL-AHO, an enhanced Archerfish Hunting Optimizer (AHO) variant that integrates Opposition-Based Learning (OBL) to improve AHO's exploratory capabilities. By leveraging opposite solutions alongside newly generated ones, OBL-AHO broadens the search space, effectively reducing the risk of local optima entrapment-a common challenge in metaheuristic algorithms. The performance of OBL-AHO is rigorously compared with the original AHO and several prominent Metaheuristic Algorithms (MAs), including TLBO (Teaching-Learning-Based Optimization), PSO (Particle Swarm Optimization), GWO (Grey Wolf Optimizer), OOBO (OnE-to-One Based Optimizer), SSA (Salp Swarm Algorithm). Evaluations were conducted using test functions taken from CEC 2022, and the outcomes were statistically analyzed using the Friedman test. The findings demonstrate that OBL-AHO significantly improves exploration while maintaining competitive exploitation relative to its opponents.