Grey wolf optimization (GWO) is a metaheuristic algorithm that is inspired by the behavior of grey wolves in leadership and hunting (Mirjalili et al., 2014). The algorithm classifies a population of possible solutions into four types of wolves α, β, δ, and ω.

In summary, the GWO algorithm has the ability to balance mining and exploration, accelerate the convergence speed in global search by linear convergence, improve the optimization accuracy in local search^38,39, and avoid the emergence of local extreme points in small regions⁴⁰.

The results showed that the proposed method is able to be very effective in training MLPs. For one, the GWO-MLP has very a high level of local optima avoidance, which enhances the probability of finding proper approximations of the optimal weights and biases for MLPs.