22AD406 ARTIFICIAL INTELLIGENCE

2 MARKS
UNIT I INTELLIGENT AGENTS AND BLIND SEARCH:

Q1: Define Artificial Intelligence.

A: Artificial Intelligence (AI) is the simulation of human intelligence in machines that can
perform tasks such as learning, problem-solving, and decision-making.

Q2: What is an Intelligent Agent?

A: An intelligent agent is an entity that perceives its environment through sensors and acts upon
it using actuators to achieve a specific goal.

Q3: Who is considered the father of AI?

A: John McCarthy is considered the father of AI, as he coined the term "Artificial Intelligence"
in 1956.

Q4: What was the significance of the Dartmouth Conference (1956) in AI?
A: The Dartmouth Conference marked the official beginning of AI as a field of study, bringing
researchers together to discuss machine intelligence.

Q5: What are the two main components of an intelligent system?

A: The two main components are the agent (which perceives and acts) and the environment
(where the agent operates).

Q6: What is a fully observable environment?

A: A fully observable environment is one where the agent has access to complete information
about the environment's state at any time.

Q7: What is a rational agent?

A: A rational agent is an entity that takes actions to maximize its performance measure based on
the given percepts and knowledge.

Q8: How is rationality different from intelligence in AI?

A: Rationality refers to making the best possible decision, while intelligence encompasses
learning, reasoning, and problem-solving.

Q9: What is a static environment in AI?

A: A static environment does not change while the agent is making a decision.

Q10: What is a stochastic environment?

A: A stochastic environment has unpredictable outcomes, meaning the same action may lead to
different results.

Q11: Name two types of intelligent agents.

A: (i) Simple Reflex Agent, (ii) Goal-Based Agent.
Q12: What is a utility-based agent?
A: A utility-based agent chooses actions that maximize its utility function, providing the best
possible outcome rather than just a goal-based result.

Q13: What is a state space in AI?

A: A state space is the set of all possible states an agent can be in while solving a problem.

Q14: What is the purpose of state space search?

A: State space search is used to explore different states systematically to find a solution to a
given problem.

Q15: What is the Generate and Test method?

A: It is a brute-force search technique where potential solutions are generated and tested to see if
they meet the goal criteria.

Q16: What is the major drawback of Generate and Test search?

A: It can be inefficient and time-consuming, as it does not use heuristics to guide the search.

Q17: What is Simple Search in AI?

A: Simple Search is an uninformed search technique that systematically explores states without
considering the cost or efficiency of paths.

Q18: Give an example of a Simple Search algorithm.

A: Breadth-First Search (BFS) is an example of a simple search algorithm.

Q19: What is Depth-First Search (DFS)?

A: DFS is a search algorithm that explores as far as possible along a branch before backtracking.

Q20: What is the worst-case time complexity of DFS?

A: The worst-case time complexity of DFS is O(b^m), where b is the branching factor and m is
the maximum depth.

Q21: What is Breadth-First Search (BFS)?

A: BFS is a search algorithm that explores all nodes at the current level before moving deeper
into the search tree.

Q22: What is the advantage of BFS over DFS?

A: BFS guarantees finding the shortest path in an unweighted graph, whereas DFS does not.

Q23: Which search method uses more memory, DFS or BFS?

A: BFS uses more memory as it stores all nodes at the current level, whereas DFS only stores the
current path.

Q24: Which search method is more suitable for finding a solution in an infinite state space?
A: DFS can get stuck in an infinite loop, while BFS systematically explores all levels, making it
more suitable for infinite state spaces.
Q25: What is Depth-Bounded DFS?
A: Depth-Bounded DFS is a variation of DFS where a depth limit is set to prevent exploring
deep or infinite paths.

Q26: How does Depth-Bounded DFS solve the problem of infinite loops in DFS?
A: By setting a maximum depth, it ensures the search does not continue indefinitely in deep or
infinite state spaces.

Q27:Define an Intelligent Agent.

Answer: An intelligent agent is an entity that perceives its environment through sensors and acts
upon it using actuators to achieve specific goals.

Q28:What are the main components of an intelligent agent?

Answer: The main components are sensors, actuators, an agent function, and a performance
measure.

Q29:What is the difference between an agent and an environment?

Answer: An agent is an entity that perceives and acts, while the environment is the external
system in which the agent operates.

Q30:Define Rationality in AI.

Answer: Rationality refers to an agent’s ability to make decisions that maximize its performance
measure based on available information and possible actions.

Q31:What is the difference between BFS and DFS?

Answer: BFS explores all nodes at the current level before moving deeper, whereas DFS
explores as far as possible along a branch before backtracking.

Q32:What is Generate and Test search?

Answer: Generate and Test is a brute-force search technique where possible solutions are
generated and tested to check if they meet the goal criteria.

Q33:What is Depth Bounded DFS?

Answer: Depth Bounded DFS is a variation of DFS where the search is limited to a fixed depth
to avoid infinite loops in deep or infinite graphs.

Q34:What is the primary advantage of BFS over DFS?

Answer: BFS guarantees finding the shortest path in an unweighted graph, whereas DFS does
not.

Q35:What is the nature of a dynamic environment?

Answer: A dynamic environment changes while the agent is deliberating on an action.

Q36:What is meant by the structure of an intelligent agent?

Answer: The structure of an intelligent agent refers to how it is designed, including its
architecture and decision-making mechanisms (e.g., simple reflex, goal-based, utility-based).
 UNIT II INFORMED SEARCH METHODS:

Q1: What is Heuristic Search?

A: Heuristic search is a search strategy that uses a heuristic function to estimate the best possible
path toward a goal, improving search efficiency.

Q2: How does heuristic search differ from uninformed search?

A: Heuristic search uses domain-specific knowledge to guide the search, while uninformed
search explores blindly without guidance.

Q3: What is a heuristic function in AI?

A: A heuristic function, denoted as h(n), estimates the cost or distance from a given node to the
goal state.

Q4: Give an example of a commonly used heuristic function.

A: The Manhattan distance (sum of horizontal and vertical distances) is a common heuristic
function used in grid-based pathfinding.

Q5: What is Best First Search?

A: Best First Search is a search algorithm that selects the next node to expand based on the
lowest heuristic cost h(n).

Q6: How does Best First Search differ from A Search?*

A: Best First Search uses only h(n), whereas A* Search uses f(n) = g(n) + h(n), considering
both path cost and heuristic estimates.

Q7: What is the Hill Climbing search algorithm?

A: Hill Climbing is an iterative search algorithm that moves toward the highest-valued
neighboring state, optimizing locally at each step.

Q8: What is the main drawback of Hill Climbing?

A: It can get stuck in local maxima, plateaus, or ridges, preventing it from reaching the global
optimum.

Q9: What is a Local Maximum in Hill Climbing?

A: A local maximum is a state where all neighboring states have lower values, but it is not the
best possible solution (global maximum).

Q10: Name one technique to overcome local maxima in Hill Climbing.

A: Random restarts, where the search is restarted from a new random position, can help escape
local maxima.

Q11: What is the Solution State Space in AI?

A: It refers to the entire set of possible states that an agent can explore while solving a problem.
Q12: What defines the goal state in a solution state space?
A: The goal state is a specific state that satisfies the problem’s objective or solution criteria.

Q13: What is Variable Neighbourhood Descent (VND)?

A: VND is an optimization technique that systematically changes the neighborhood structure to
escape local optima.

Q14: What is the main advantage of VND?

A: It avoids getting stuck in poor local optima by exploring multiple neighborhood structures.

Q15: What is Beam Search?

A: Beam Search is a heuristic search algorithm that keeps only the k best nodes at each level to
limit memory usage.

Q16: How does Beam Search differ from Best First Search?
A: Beam Search restricts the number of nodes explored at each step, while Best First Search
expands all promising nodes.

Q17: What is Taboo Search?

A: Taboo Search is a metaheuristic optimization technique that prevents revisiting recently
explored states using a taboo list.

Q18: Why is the taboo list important in Taboo Search?

A: It helps escape cycles and local optima by preventing the algorithm from revisiting previously
explored solutions.

Q19: What is the Peak to Peak method in AI?

A: Peak to Peak is an optimization technique that allows movement between local maxima using
a special transition strategy.

Q20: How does Peak to Peak search overcome local maxima?

A: It introduces strategies like random jumps or guided transitions to move between peaks (local
optima).

Q21: What is Brute Force Search?

A: Brute Force Search systematically explores all possible solutions until the correct one is
found.

Q22: What is the main disadvantage of Brute Force Search?

A: It is highly inefficient for large problems due to its exponential time complexity.

Q23: What is the Branch and Bound method?

A: Branch and Bound is an optimization algorithm that systematically explores branches of a
search tree while pruning non-optimal solutions.
Q24: How does Branch and Bound improve search efficiency?
A: It eliminates unnecessary branches using bounding functions, reducing the number of
explored states.

Q25: What is Refinement Search?

A: Refinement Search incrementally improves a partial solution until a complete and optimal
solution is found.

Q26: What is an example of Refinement Search?

A: Constraint Satisfaction Problems (CSPs), where a solution is refined step-by-step by adjusting
variable assignments.

UNIT III A* AND RANDOMIZED SEARCH METHODS

Q1: What is the A algorithm?*

A: A* is a best-first search algorithm that finds the shortest path using the evaluation function
f(n) = g(n) + h(n), where g(n) is the actual cost and h(n) is the heuristic estimate to the goal.

Q2: Why is A better than BFS and DFS?*

A: A* efficiently balances exploration and cost optimization, ensuring it finds the shortest path if
the heuristic function is admissible.

Q3: When is A considered admissible?*

A: A* is admissible if the heuristic function h(n) never overestimates the actual cost to the goal
(i.e., h(n) is optimistic).

Q4: Why does A need an admissible heuristic?*

A: An admissible heuristic ensures A* always finds the optimal path, avoiding suboptimal
solutions.

Q5: What is Recursive Best First Search (RBFS)?

A: RBFS is a memory-efficient version of A* that uses recursion to store only limited paths
instead of keeping all nodes in memory.

Q6: How does RBFS handle limited memory constraints?

A: RBFS replaces the least promising nodes with better alternatives, reducing memory usage
compared to standard A*.

Q7: What is a local maximum in search algorithms?

A: A local maximum is a suboptimal peak where all neighboring solutions have lower values,
preventing further improvement.

Q8: Name two methods to escape local maxima.

A: Simulated Annealing and Iterated Hill Climbing.
Q9: What is Iterated Hill Climbing?
A: Iterated Hill Climbing repeatedly runs Hill Climbing with different random starting points to
escape local maxima.

Q10: How does Iterated Hill Climbing improve standard Hill Climbing?
A: It avoids getting stuck in local maxima by restarting the search from different initial positions.

Q11: What is Simulated Annealing in AI?

A: Simulated Annealing is a probabilistic search method that allows occasional worse moves to
escape local maxima.

Q12: What is the role of temperature in Simulated Annealing?

A: Temperature controls the probability of accepting worse solutions, gradually decreasing over
time to refine the search.

Q13: What is a Genetic Algorithm?

A: A Genetic Algorithm (GA) is an optimization technique inspired by natural selection, using
selection, crossover, and mutation to evolve solutions.

Q14: Name three key operations in Genetic Algorithms.

A: Selection, Crossover, and Mutation.

Q15: What is the Travelling Salesman Problem (TSP)?

A: The TSP is an optimization problem where a salesman must visit N cities exactly once and
return to the starting point while minimizing travel cost.

Q16: Why is TSP considered an NP-hard problem?

A: The number of possible routes grows exponentially with the number of cities, making it
computationally expensive to find the optimal solution.

Q17: How can Genetic Algorithms be used to solve TSP?

A: GA can optimize TSP by evolving better tour sequences through selection, crossover, and
mutation over multiple generations.

Q18: What type of crossover is commonly used in GA for TSP?

A: Order Crossover (OX) and Partially Mapped Crossover (PMX) are commonly used to
maintain valid city sequences.

UNIT IV GAME PLAYING, PLANNING AND CONSTRAINT SATISFACTION

Q1: What is an example of an AI-based board game?

A: Chess, Go, and Tic-Tac-Toe are examples of board games where AI is used to make
intelligent moves.
Q2: How does AI play board games?
A: AI plays board games using search algorithms like Minimax, Alpha-Beta pruning, and Monte
Carlo Tree Search (MCTS).

Minimax Algorithm

Q3: What is the Minimax algorithm?

A: Minimax is a decision-making algorithm used in two-player games to minimize the
opponent’s maximum possible gain.

Q4: What is the main assumption behind the Minimax algorithm?

A: It assumes that both players play optimally and try to maximize their own chances of
winning.

Alpha-Beta Pruning

Q5: What is Alpha-Beta pruning?

A: Alpha-Beta pruning is an optimization technique for Minimax that eliminates unnecessary
branches, reducing computation time.

Q6: What is the advantage of Alpha-Beta pruning over Minimax?

A: It speeds up the Minimax algorithm by ignoring branches that will not affect the final
decision.

B Search*

Q7: What is B Search?*

A: B* Search is an extension of Alpha-Beta pruning that uses probability to decide which
branches to explore.

Q8: When is B Search useful?*

A: It is useful when the search space is large, and exact evaluation of all possible moves is
impractical.

Q9: What is the major limitation of search-based game-playing AI?

A: Search algorithms become computationally expensive as the game complexity increases.

Q10: How can heuristic functions help overcome search limitations?

A: Heuristics guide the search toward promising moves, reducing the number of nodes explored.

Q11: What does STRIPS stand for in AI?

A: STRIPS stands for Stanford Research Institute Problem Solver, a formal planning system.

Q12: What is the purpose of STRIPS in AI?

A: STRIPS represents planning problems using initial states, goal states, and operators to
achieve goals.
Q13: What is Forward State Space Planning?
A: It is a planning method that starts from the initial state and applies actions until the goal is
reached.

Q14: What is the main drawback of Forward State Space Planning?

A: It can be computationally expensive because it explores many possible future states.

Q15: What is Backward State Space Planning?

A: It starts from the goal state and works backward to find a sequence of actions leading to the
initial state.

Q16: When is Backward State Space Planning more efficient than Forward Planning?
A: When the number of backward possibilities is smaller than forward possibilities, reducing the
search space.

Q17: What is Goal Stack Planning?

A: Goal Stack Planning is a problem-solving approach where goals are pushed onto a stack and
solved sequentially.

Q18: What is a drawback of Goal Stack Planning?

A: It can lead to inefficient solutions if solving one goal undoes progress on another goal.

Q19: What is a Constraint Satisfaction Problem (CSP)?

A: A CSP is a problem where a solution must satisfy a set of constraints (e.g., Sudoku, Map
Coloring).

Q20: What are the three main components of a CSP?

A: Variables, Domains, and Constraints.

Q21: What is the N-Queens Problem?

A: The N-Queens problem requires placing N queens on an N×N chessboard so that no two
queens attack each other.

Q22: What is one approach to solving the N-Queens problem?

A: Backtracking is a common approach used to solve the N-Queens problem.

UNIT V PREPOSITIONAL LOGIC, FIRST ORDER LOGIC AND INFERENCING

Q1: What is formal logic in AI?

A: Formal logic is a system of reasoning using symbols and rules to represent and infer
knowledge systematically.

Q2: Name two types of formal logic used in AI.

A: Propositional Logic (PL) and First-Order Logic (FOL).
Q3: What is Propositional Logic (PL)?
A: PL is a logic system where statements (propositions) are either true or false, combined using
logical operators like AND (∧), OR (∨), and NOT (¬).

Q4: What is a truth table in Propositional Logic?

A: A truth table shows all possible truth values of a logical expression based on its inputs.

Q5: What is resolution in Propositional Logic?

A: Resolution is a rule of inference used to prove logical statements by eliminating
contradictions.

Q6: What is the purpose of the resolution rule?

A: It helps in automated theorem proving by deducing new clauses from existing ones.

A: FOL extends Propositional Logic by introducing quantifiers (∀, ∃) and predicates to express
Q7: What is First-Order Logic (FOL)?

complex relationships.

Q8: How does FOL differ from Propositional Logic?

A: FOL includes objects, predicates, and quantifiers, whereas Propositional Logic deals only
with simple statements.

Q9: What is Forward Chaining?

A: Forward Chaining starts from known facts and applies inference rules to derive new facts
until a goal is reached.

Q10: Why is Forward Chaining incomplete in FOL?

A: Because it may generate an infinite number of conclusions, missing the most relevant ones
needed for the solution.

Q11: What is Resolution Refutation in FOL?

A: It is a proof technique that assumes the negation of a statement and derives a contradiction to
prove the original statement.

Q12: How does Resolution Refutation help in theorem proving?

A: It systematically eliminates clauses until a contradiction is found, proving the statement by
contradiction.

Q13: What is a Horn Clause in logic?

A: A Horn Clause is a disjunction of literals with at most one positive literal, commonly used in
logic programming.

Q14: What is SLD (Selective Linear Definite) Resolution?

A: SLD Resolution is a special case of resolution used in logic programming for efficient proof
search in Prolog.
Q15: What is Backward Chaining?
A: Backward Chaining starts from the goal and works backward by finding rules that support the
goal until known facts are reached.

Q16: How does Backward Chaining differ from Forward Chaining?

A: Backward Chaining starts from the goal and works backward, whereas Forward Chaining
starts from facts and moves forward.