Comparison of Bayesian Optimization Algorithms for BBOB Problems in Dimensions 10 and 60
Maria Laura Santoni, Elena Raponi, Renato De Leone, Carola Doerr · 2023
Bayesian Optimization (BO) is a class of black-box, surrogate-based heuristics that can efficiently optimize problems that are expensive to evaluate and therefore allow only small evaluation budgets. Regardless of the size of the budget, high dimensionality also poses a challenge to BO, whose performance reportedly often suffers when the dimension exceeds 15 variables. Many new algorithms have been proposed to address this problem. However, it is not well understood which one is the best for which optimization scenario.