Benchmarking the SMS-EMOA with self-adaptation on the bbob-biobj test suite
Simon Wessing · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2017
Variation operators have seemingly been less in the focus than selection operators during the first years of research on evolutionary multiobjective optimization. Several new developments in benchmarking and hypervolume selection have now sparked a renewed interest in the topic. Here, we benchmark a variant of the S-metric selection evolutionary multi-objective optimization algorithm with self-adaptive mutative control of a single step size parameter, but without recombination. It obtains better results than variants with differential evolution or polynomial mutation as variation.