At a first look, q-learning is a revolutionary strategy in realizing Artificial Intelligence. It has to do with finding a policy, a reward structure, neural networks for storing the q-table and a markov decision process. Introducing all the terms is an interesting task and it's a lot of fun to explain the details to the beginner.
On the other hand, there is a well known paradigm available, which describes how games are played by an artificial intelligence. Under the broad term “gametree” the overall time period is described, in which actions in the game took place. It means, that a player can make a move, and in reaction the other player can take a step too. The game log is stored in a hierarchical structure which gives an overview of potential alternatives in a game. It make sense to ask for a certain node in the gametree what the perfect move is.
From the perspective of teaching computer science the question is, how to introduce qlearning from the perspective of a gametree. That means, if a reinforcement learning algorithm gets started, the computer will search in the game tree. Is he doing so?