Which measure evaluates how good a specific state is, in terms of expected future rewards?

Which measure evaluates how good a specific state is, in terms of expected future rewards?

• A. Action-Value Function
• B. Reward
• C. Value Function
• D. State

Answer: (C)

The key idea is measuring how good it is to be in a particular state based on the expected future rewards you can obtain from that state. This is captured by the value function. The value function assigns a number to each state that reflects the expected return (the sum of future rewards, typically discounted) you'd receive starting from that state while following a given policy. It tells you how desirable it is to be in that state in the long run, not just the immediate payoff. This differs from the reward, which is the immediate payoff you get at a transition; from the action-value function, which evaluates the quality of taking a specific action in a state; and from the state itself, which is just the situation rather than its desirability. Hence, the measure described is the value function.

The key idea is measuring how good it is to be in a particular state based on the expected future rewards you can obtain from that state. This is captured by the value function. The value function assigns a number to each state that reflects the expected return (the sum of future rewards, typically discounted) you'd receive starting from that state while following a given policy. It tells you how desirable it is to be in that state in the long run, not just the immediate payoff.

This differs from the reward, which is the immediate payoff you get at a transition; from the action-value function, which evaluates the quality of taking a specific action in a state; and from the state itself, which is just the situation rather than its desirability. Hence, the measure described is the value function.