Improving Cuckoo Search Algorithm With Mittag-Leffler Distribution

Cuckoo search (CS), as one of the recent nature-inspired metaheuristic algorithms, has proved to be an efficient approach due to the combination of Lévy flights, local search capabilities and guaranteed global convergence. CS uses Lévy flights in global random walk to explore the search space. The Lévy step is taken from the Lévy distribution which is a heavy-tailed probability distribution. In this case, a fraction of large steps are generated, which plays an important role in enhancing search capability of CS. Besides, although many foragers and wandering animals have been shown to follow a Lévy distribution of steps, investigation into the impact of other different heavy-tailed probability distributions on CS is still insufficient up to now. Based on the above considerations, we are motivated to apply the well-known Mittag-Leffler distribution to the standard CS algorithm, and proposed an improved cuckoo search algorithm (CSML) in this paper, where a more efficient search is supposed to take place in the search space thanks to the long jumps. In order to verify the performance of CSML, experiments are carried out on a test suite of 20 benchmark functions. In terms of the observations and results analysis, CSML can be regarded as a new potentially promising algorithm for solving optimization problems.