Question

1.Why does a search in a game return a strategy rather than a solution? (Check all that apply

		Most games have multiple solutions, so a single solution cannot be returned.
		A game agent will typically have its options limited by its opponent, since adversaries in a game tend to have opposing goals.
		All games are zero-sum games, thereby necessitating a strategy instead of a solution.
		A game agent's choice of action depends on what its opponent does, so a solution that is available earlier in a game may not be available later in the game 2.Which of the following are true regarding the minimax algorithm?
It uses a max-value function that computes the maximum of the minimum values chosen by MIN.
		The minimax values are always computed beginning at the terminal states and progressing toward the initial state.
		The minimax values that are calculated always refer to the utility of a state with respect to MAX.
		All of the above

3.For a utility function used in a game search, which state or states are most key to the value the function returns?

the initial state only
		the terminal state only
		both the initial and final states
		all states from the initial state (inclusive) to the final state (inclusive)

3 part 2. A state that has a drastically different value, according to the evaluation function, from the states that closely precede it or closely succeed it (or both) in a game tree, is known as what type of state

anomalous
		non-quiescent
		volatile
		fluctuating

Answer 1

1.Why does a search in a game return a strategy rather than a solution?

A game agent's choice of action depends on what its opponent does, so a solution that is available earlier in a game may not be available later in the game.

=> because strategy helps in deciding the move for every possible opposite player reply.

2.Which of the following are true regarding the minimax algorithm?

The minimax values are always computed beginning at the terminal states and progressing toward the initial state.

3.For a utility function used in a game search, which state or states are most key to the value the function returns?

all states from the initial state (inclusive) to the final state (inclusive)

=> all terminal states will be having a numeric value and utility function is applied to the leaf nodes and the values are backed up (for a max-position: back up the value of largest successor; of a min-position: back up the value of smallest successor). During the minimax procedure, starting from leaf nodes we back up values all the way up to root node. m level apply the utility function, back-up values all the way up to the root node, and that node selects the move.

3 part 2. A state that has a drastically different value, according to the evaluation function, from the states that closely precede it or closely succeed it (or both) in a game tree, is known as what type of state

non-quiescent