MS&E338 Reinforcement Learning


rat in maze with gps 


Can reinforcement learning serve as a basis for artificial general intelligence? This course offers an advanced introduction to reinforcement learning, oriented around research topics that aim to advance the field toward that goal. This year's offering will emphasize value function learning, exploration, state abstraction, and general value functions. Content will focus on general concepts, algorithms, and mathematical analysis. Students are expected to have experience with optimization, stochastic processes, supervised learning, numerical computation, and mathematical proofs.


‘‘Other areas of machine learning are about minimization; reinforcement learning is about maximization.’’