“…An important and classical aspect of EAs is how robust their performance is in the presence of noise [15,8]. This theme has gained increased attention in the last few years, [16,29,9,3,5,4,1,27,34,33,20], see [35] for a comprehensive review. Mostly, noise is modeled by imperfect fitness function evaluations thatinstead of the exact fitness value -return a perturbed value (e.g., by a Gaussian additive term).…”