Q34: (15 points) What are the two key steps of value learn…
Q34: (15 points) What are the two key steps of value learning? (4 points) What are the two key steps of policy gradient? (4 points) If we are to build a reinforcement learning with discrete actions, which method we should use? (2 points) What are the weaknesses of value learning (e.g., DQN)? (5 points)