A First Mathematical Runtime Analysis of the Non-dominated Sorting Genetic Algorithm II (NSGA-II)

The non-dominated sorting genetic algorithm II (NSGA-II) is the most intensively used multi-objective evolutionary (MOEA) in real-world applications. However, contrast to several simple MOEAs analyzed also via mathematical means, no such study exists for NSGA-II so far. In this work, we show that runtime analyses are feasible NSGA-II. As particular results, prove with a population size larger than Pareto front by constant factor, two classic mutation operators and three different ways select parents satisfies same asymptotic guarantees as SEMO GSEMO algorithms on basic OneMinMax LOTZ benchmark functions. if only equal of front, then cannot efficiently compute full (for an exponential number iterations, will always miss fraction front). Our experiments confirm above findings.