Search

In numerous real-life applications, nature-inspired population-based search algorithms have been applied to solve numerical optimization problems. This paper focuses on a simple and powerful swarm optimizer, named Wild Geese Algorithm (WGA), for large-scale global optimization whose efficiency and performance are verified using large-scale test functions of IEEE CEC 2008 and CEC 2010 special sessions with high dimensions D ​= ​100, 500, 1000. WGA is inspired by wild geese in nature and models various aspects of their life such as evolution, regular cooperative migration, and fatality. The effectiveness of WGA for finding the global optimal solutions of high-dimensional optimization problems is compared with that of other methods reported in the previous literature. Experimental results show that the proposed WGA has an efficient performance in solving a range of large-scale optimization problems, making it highly competitive among other large-scale optimization algorithms despite its simpler structure and easier implementation. The source code of the proposed WGA algorithm is publicly available at github.com/ebrahimakbary/WGA.