Akademik Veri Yönetim Sistemi
11th Annual International Conference on Industrial Engineering and Operations Management, IEOM 2021, Virtual, Online, 7 - 11 Mart 2021, ss.2975-2982, (Tam Metin Bildiri)
Metaheuristic methods have many design parameters, and fine-tuning of that parameters can improve these algorithms performance. In this paper, a sampling based algorithm configuration approach is proposed and applied it to the Differential Ant-Stigmergy Algorithm (DASA)’s five control parameters. Performance of a large parameter set of DASA, obtained by Latin Hypercube Hammersley Sampling (LHHS) method, used to solve the Sphere function and compared it with another tuned version. The results of our experiment demonstrated that LHHS method found better performing configurations than the default parameter value of the DASA and also than another proposed tuned version DASA*. And the results demonstrated that three parameter configurations obtained with LHHS found better result than the best configuration obtained with Sobol Sequence Sampling method (DASA*) for function dimension 20, and five parameter configuration for function dimension 40. According to the results, it can be said that usage of LHHS for initialization of other state-of-art algorithm configuration methods instead of other sampling methods is worth investigating.