Model-based algorithms: Model-based algorithms take a different approach to reinforcement learning. Instead of evaluating the value of states and actions, they try to predict the state of the ...