שאלה | 1. Which of the following statements is correct with regard to model-based RL:

1. Which of the following statements is correct with regard to model-based RL:

		a. Model-based RL generally requires a larger number of training samples than off-policy approaches
		b. Because of the need to model the real-world dynamics, model-based RL methods must wait until the end of the trajectory to update their model parameters
		c. Human feedback is key for the success of model-based RL, since obtaining the “ground truth” is impossible otherwise
		d. In model-based RL we try to learn both the policy and the dynamics of the environment