Fifty Years of RL in Games

Many researchers have advocated game domains as highly useful testbeds where one can cleanly isolate and study important issues faced by RL and more general AI methods in tackling messy real-world pro