Dynamic Programming — Policy Iteration & Value Iteration

Dynamic programming is RL with cheating. You already know the transition and reward functions; you just iterate the Bellman equation until V or π stops moving. It is the benchmark every sampling-based method tries to approach.