A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
Distribution of the number of citations over years.