WorksheetsAI Foundations Quiz
Total questions: 24
Worksheet time: 12mins
Which of the following most accurately differentiates a rational agent from an intelligent agent?
Rational agents always maximize performance measures, whereas intelligent agents may act sub-optimally due to bounded rationality.
Intelligent agents always maximize performance, whereas rational agents may act heuristically.
Rational agents are reactive, while intelligent agents are proactive.
Intelligent agents are hardware implementations, while rational agents are purely conceptual.
Which foundation of AI directly addresses the representation of "knowledge about knowledge" (meta-reasoning)?
Cognitive Science
Control Theory
Epistemology in Philosophy
Statistical Decision Theory
Consider an AI system applied to autonomous medical diagnosis. Which sub-area of AI is least relevant to the system's core decision-making process?
Natural Language Processing
Expert Systems
Computer Vision
Robotics Kinematics
Which of the following applications best demonstrates the integration of all four foundations of AI (Philosophy, Mathematics, Economics, Neuroscience)?
Stock Market Prediction using Deep Learning
Autonomous Vehicle Navigation
Face Recognition in Smartphones
Email Spam Filtering
In state-space search, which characteristic distinguishes control strategies from problem characteristics?
Control strategies define how states are represented, while problem characteristics define how search is expanded.
Control strategies determine node expansion order, while problem characteristics describe structure of the problem itself.
Control strategies define the branching factor, while problem characteristics define goal states.
Both are interchangeable in practical implementations.
Which search algorithm guarantees finding an optimal solution without knowing path costs?
Breadth-First Search
Depth-First Search
Branch and Bound
Iterative Deepening A*
In Bidirectional Search, which major limitation makes it impractical in many real-world problems?
Memory requirements grow quadratically with branching factor.
Both forward and backward searches must know the exact middle point in advance.
The backward search requires goal state specifications that may be difficult to reverse-engineer.
It cannot handle weighted edges in the search space.
Consider a problem where solution depth is unknown and state space is infinite. Which search strategy is most appropriate?
Depth-First Search
Breadth-First Search
Iterative Deepening Depth-First Search (DFID)
Branch and Bound
In A* search, which condition ensures optimality?
The heuristic function must dominate actual path cost.
The heuristic function must be both admissible and consistent.
The branching factor must remain constant.
The search tree must be finite.
Hill Climbing often suffers from local maxima. Which variation directly mitigates this issue by probabilistic acceptance of worse states?
Stochastic Hill Climbing
Beam Search
Simulated Annealing
Best-First Search
Why is Iterative Deepening A* (IDA*) more memory efficient than A*?
It stores only the current path instead of the entire open list.
It prunes all non-optimal branches in advance.
It uses greedy heuristics instead of admissible ones.
It does not expand nodes beyond a fixed depth.
Beam search is considered incomplete in some cases. Why?
It discards all but a limited set of best nodes at each level, potentially eliminating the optimal path.
It uses inconsistent heuristics.
It fails when branching factor is too small.
It always prefers depth over breadth.
A robot must plan a path in an unknown maze where the depth of the goal state is unknown, but node expansion is costly. Which strategy minimizes both time and memory?
Breadth-First Search
Iterative Deepening DFS
Depth-Limited Search with large cutoff
A* with admissible heuristic
If a heuristic h1 is admissible and another heuristic h2 dominates h1 (i.e., h2(n) ≥ h1(n) for all nodes), what can we conclude about A* using h2?
It expands fewer or equal nodes than h1.
It expands more nodes than h1.
It becomes incomplete.
It cannot guarantee optimality.
Which statement about A* search time and space complexity is correct?
Both time and space complexity are polynomial in the solution depth.
Time complexity is exponential, and space complexity is exponential in solution depth.
Time is polynomial, but space is exponential.
Both are linear in branching factor.
Consider two admissible heuristics h1 and h2 for the same problem. If h2(n) ≥ h1(n) for all nodes, what can be said about A* performance using h2 compared to h1?
Expands more nodes.
Expands fewer or equal nodes.
Expands exactly the same nodes.
Expands fewer nodes but loses optimality.
Which of the following modifications makes A* incomplete?
Allowing inconsistent heuristic.
Imposing a cutoff on open list size (memory-bounded A*).
Using tie-breaking in favor of deeper nodes.
Using an admissible heuristic.
Which of the following best explains why A* is not commonly used in very large state spaces (e.g., pathfinding in huge graphs)?
Lack of optimality.
Excessive re-expansion of nodes.
Exponential memory usage due to storing all frontier nodes.
Inability to handle weighted edges.
You are searching in a graph with the following edge costs: A→B = 2, A→C = 5, B→D = 4, C→D = 1. What is the cost of the optimal path from A to D found by Uniform Cost Search?
5
6
7
9
In a graph, g(n) is the cost so far, h(n) is heuristic. Suppose at node N: g(N) = 4 h(N) = 5 Actual optimal cost-to-go h*(N) = 7 Which statement is true?
h is admissible and consistent.
h is admissible but not consistent.
h is not admissible.
h is overestimating, so A* may fail.
You are solving TSP using Branch and Bound. Partial tour cost from A→B→C = 15, and the lower bound estimate from C to cover remaining cities = 20. What is the bound for this partial path?
15
20
35
Cannot be determined without full expansion.
In a problem with solution depth = 4 and branching factor = 3, how many total nodes will Iterative Deepening DFS expand (worst case)?
121
109
142
130
Consider nodes with costs: Node X: g(X)=3, h(X)=6 Node Y: g(Y)=5, h(Y)=3 Which node will be chosen first by: Greedy Best-First A*
Both choose X
Both choose Y
Greedy chooses X, A* chooses Y
Greedy chooses Y, A* chooses X
Which of the following problems is best modeled as an AND-OR search rather than a simple state-space search?
Solving 8-puzzle
Navigating a maze with multiple paths
Diagnosing a disease with multiple possible causes and treatments
Pathfinding in a weighted graph
