We said reinforcement learning was the interaction between an agent and the environment. The agent is a decision maker, it has a set of actions, a ∈ A(s), that depend of the state, s ∈ S, of the environment.

L 1 A 3 B 5 Y ? R I 3 N 5 T 1 H L 3 A 5 B 1 Y R Y 1 N 5 T 3 H L 5 A 2 B Y R ? Y N 2 T 5 H L 1 A 4 B Y ? R Y N 1 T 4 H L 2 A B Y R ? Y N T 2 H L 4 A B Y ? R Y N T 4 H L A B Y R ? Y N T H L A B Y R Y N T