Self-organizing migrating algorithm with clustering-aided migration
Tomáš Kadavý, Michal Pluháček, Adam Viktorin, Roman Šenkeřík · 2020
This paper proposes a novel migration strategy for Self-organizing Migrating Algorithm (SOMA), which combines advantages of the explorative All-To-Random migration with new exploitation focused All-To-Cluster-Leaders strategy. The main goal of this novel innovation to SOMA is to deliver competitive results, not only on the latest CEC 2020 benchmark set on a single objective bound-constrained numerical optimization. The proposed algorithm variant was titled SOMA-CL, and it has manifested notable potential in such demanding challenges. The results of the proposed algorithm were compared and tested for statistical significance against two other SOMA variants.