In this paper, we study the pure exploration model with general distribution functions, which means that the reward function of each arm depends on the whole distribution, not only its mean. We adapt ...