Unit 3 GAME PLAYING AND PLANNING Flashcards
Q: What is the definition of a game in AI?
A: A game in AI is defined as a scenario within a multiagent environment where agents interact according to rules that define permissible actions and outcomes.
Q: What are the components of a game in AI?
A: The components include Initial State (S0), PLAYER(s), ACTIONS(s), Transition Model (RESULT(s,a)), Terminal Test (TERMINAL-TEST(s)), and Utility Function (UTILITY(s, p)).
Q: What is the Minimax Algorithm used for in AI?
A: The Minimax Algorithm is used for making optimal decisions in adversarial settings by evaluating future states to determine the best possible move for a player.
Q: What is Alpha-Beta Pruning?
A: Alpha-Beta Pruning is an optimization technique used with the Minimax Algorithm to reduce the number of nodes evaluated, making the search more efficient.
Q: What is the difference between deterministic and stochastic games?
A: In deterministic games, outcomes are determined solely by players’ actions, while stochastic games involve random elements, such as dice rolls or card draws.
Q: How does the Expectiminimax Algorithm work?
A: The Expectiminimax Algorithm extends Minimax for stochastic games by calculating the expected value of positions, averaging outcomes based on their probabilities.
Q: What is imperfect real-time decision making in games?
A: It refers to making quick decisions in real-time games despite incomplete information, using heuristic evaluation functions or Monte Carlo simulations.
Q: What is a game tree in AI?
A: A game tree is a graphical representation of all possible moves and outcomes in a turn-based game, with nodes representing game states and edges representing player moves.
Q: What is the role of chance nodes in stochastic games?
A: Chance nodes represent random events, such as dice rolls, that influence the possible actions and outcomes in a game.
Q: What is the significance of the utility function in AI games?
A: The utility function measures the desirability of game outcomes, helping to rank game states based on their benefit to the player.
Q: What is Goal Stack Planning in AI?
A: Goal Stack Planning is a type of planning system where goals are ordered in a stack, and the system tries to achieve the topmost goal, adjusting the plan based on achieved subgoals.
Q: What is the difference between a single-agent problem and a multi-agent problem in AI?
A: A single-agent problem involves only one agent making decisions, while a multi-agent problem involves multiple agents, where each agent’s action can affect the other agents’ outcomes.
Q: What is a heuristic function in game-playing algorithms?
A: A heuristic function is an evaluation function used to estimate the value or utility of a game state, often guiding decisions when searching through possible moves.
Q: What is Monte Carlo Tree Search (MCTS) in AI?
A: MCTS is a heuristic search algorithm used for decision-making in games that uses random sampling of the search space to estimate the best move.
Q: What is a zero-sum game?
A: A zero-sum game is a game where one player’s gain is exactly balanced by the loss of another player, making the total utility constant.
