Sampling Effects on Algorithm Selection for Continuous Black-Box Optimization

In this paper, we investigate how systemic errors due to random sampling impact on automated algorithm selection for bound-constrained, single-objective, continuous black-box optimization. We construct a machine learning-based selector, which uses exploratory landscape analysis features as inputs. test the accuracy of recommendations experimentally using resampling techniques and hold-one-instance-out hold-one-problem-out validation methods. The results demonstrate that selector remains accurate even with noise, although not without trade-offs.