שאלה | 10. [AlphaGo/Zero] Which of the following statements is true regarding AlphaGo and AlphaZero:(23)

10. [AlphaGo/Zero] Which of the following statements is true regarding AlphaGo and AlphaZero:(23)

		a) Unlike AlphaGo, AlphaZero does not use MCTS.
		b) The AlphaZero network outputs both the Q-values and the Value function (i.e., the probability of winning). The former is used to select the action, the latter is used for navigating the branches of the MCTS.
		c) Both architectures use self-play to improve their performance.
		d) Both approaches use the same state-space representation but differ in their architectures and training processes.