Investigation on parameterless schemes for DSMGA-II
Journal
GECCO 2016 Companion - Proceedings of the 2016 Genetic and Evolutionary Computation Conference
Pages
85-86
Date Issued
2016
Author(s)
Chang, C.-H.
Abstract
The only parameter that dependency structure matrix genetic algorithm II (DSMGA-II) requires is the population size, and the practicability of DSMGA-II would be further enhanced by removing the parameter. Existing parameterless schemes cannot be directly applied to DSMGA-II due to the confliction with the back mixing---one of the major operators of DSMGA-II. This paper focused on developing such parameterless schemes for DSMGA-II. Empirically these scheme yields promising results.
SDGs
Type
conference paper