The Reinforcement Learning Problem | part of Reinforcement Learning: An Introduction | MIT Press books | IEEE Xplore