Which of the following is assumed to be known in a Markov Decision Process but not in a reinforcement learning problem?
a) Transition Probabilities
b) Rewards
c) State Space
d) Optimal Policy