Reinforcement learning is a type of artificial intelligence and machine learning algorithm that finds the optimal decision-making policy by using experiences in an unknown environment. It mimics the learning methods of animals by rewarding desired behaviors, encouraging the repetition of those behaviors as a way to learn.
As seen in the video above, the reinforcement learning agent initially does not know the rules of the game, but as it plays, it learns how to play. Eventually, it becomes better at the game than a human.Â