Train agents through rewards, policies, value functions, exploration, exploitation, and simulation environments. You will apply reinforcement learning to small control and decision problems before judging where it is practical.