Shift-invariant grey wolf optimizer exploiting reference points and random selection of step-sizes
Keiji Tatsumi, Nao Kinoshita · 2022 61st Annual Conference of the Society of Instrument and Control Engineers (SICE) · 2022
The grey wolf optimizer (GWO) is a popular metaheuristic method based on the leadership hierarchy of grey wolves to solve continuous optimization problems, which has been received a lot of attention from researchers because of its advantages over other metaheuristic methods and its simplicity. In this paper, we empirically and theoretically show that the method is shift-dependent on the coordinate representation of the optimization problem, and in addition, it is too specialized only to optimization problems having the optimal solution at the origin, which means that its searching ability is not necessarily effective to other problems. Then, we propose a new improved GWO which is invariant to the shift-transformation by using reference points and step-size selection based on a random permutation in its updating system. Finally, we show advantages of the proposed GWO by comparing it with other metaheuristic methods.