Midterm Flashcards
In the game of chance, when adding the chance node between MAX and MIN steps, how can we get the value at MAX step on the top, given the values of the lower layer?
a) By card shuffing
b) By tossing a coin
c) By choosing the maximum value
d) By checking alpha and beta
e) By rolling a dice
c) By choosing the max value
The Euclidean distance is an admissible heuristic for Pacman path-planning problems
True/False
True
Given two admissible heuristics h1 and h2. Which one of the following is guaranteed to also be admissible heuristic?
a) h1 * h2
b) 0.4h1 + 0.6h2
c) 0.5h1 + 1.5h2
d) 2 * min(h1,h2)
b) 0.4h1 + 0.6h2
What is CORRECT about the semi-dynamic property of the environment?
a) The current state or outcome could effect all future states
b) The next state is determined by the current state and the action executed by the agent
c) The agent has access to the complete state of the environment at each point in time
d) As time progress the environment does not change, but passage of time will affect the performance of the agent
e) At the moment that the agent calculated the decision, the environment does not change.
d) As time progress, the environment does not change, but passage of time will effect the performance of the agent.
What is correct about the minimax algorithm?
a) Minimize our utility, and maximize opponent’s utility
b) If the tree is infinite, minimax algorithm always finds a solution of one exists.
c) Based on breadth-first search
d) Achieve good performance in the game with limited time playing.
e) Maximize our utility, and minimize opponent’s utility.
e) Max our’s, min their’s
Which of the following statement is TRUE
a) Hill-Climbing algorithm is always stuck in the local maxima
b) Random starts are often used in local search to diminish the problem of local maxima
c) Simulated annealing ensures that you will always reach a global optimum
b) Random starts are often used in local search to diminish the problem of local maxima
In alpha-beta pruning, which statement is CORRECT?
a) Beta is the best value to MAX found so far off the current path
b) Alpha is the best value to MIN found so far off the current path
c) In the MIN step, only accept if the proposed value is greater than beta
d) Only update the current value if the proposed value is better regarding the meaning of the current step (MAX/MIN)
e) Alpha, initiated with +inf, propagated upward.
d) Only update the current value if the proposed value is better regarding the meaning of the current step (MAX/MIN)
The contingency problem describes which type of environment? Choose the MOST CORRECT
a) Unknown state space env.
b) Non-deterministic and/or partially observable env.
c) Deterministic and fully observable env.
d) None of the answer is correct
e) Non-observable env.
b) Non-deterministic and/or partially observable env.
Which is the role of the problem generator in a goal-based agent?
a) The problem generator determines how to improve the performance element in the future
b) There is no problem in this agent type
c) The problem generator is responsible to select the external actions
d) The problem generator suggests some extra actions for the performance element to try
e) The problem generates the feedback on how good or bad the agent performed in the previous moment
b) There is no problem generator in this agent type
Which is TRUE
a) The sum of several admissible heuristics is still and admissible heuristic.
b) A* search with a heuristic that is not completely afmissible may still find the shorted path to the goal state
c) To guarantee the optimality of A* search, the heuristic function must overstimate the actual cost to reach the goal state.
b) A* search with a heuristic that is not completely afmissible may still find the shorted path to the goal state
