Roulette-wheel selection is a frequently used method in genetic and evolutionary algorithms or in modeling of complicated networks. Existing routines select considered one of N individuals using 온라인 슬롯 머신 search algorithms of O or O(log) complexity. We current a easy roulette-wheel selection algorithm, which usually has O complexity and is predicated on stochastic acceptance instead of looking. We also discuss a hybrid model, which might be appropriate for extremely heterogeneous weight distributions, discovered, for example, in some models of complicated networks. With minor modifications, the algorithm might also be used for sampling with health cut-off at a sure worth or for sampling with out replacement.