UNIT 3 and 4
1. Define Unification.
Answer:
Unification in AI is the process of making two logical expressions identical by finding suitable substitutions for variables. It is mainly used in first-order logic and automated reasoning systems.
Parent(X, Mary) and Parent(John, Mary)
If we substitute X = John, both expressions become identical.
🏷 Not for Examination – Only for Understanding Purpose
2. State the conditions for successful unification.
Answer:
Unification is successful when identical constants match, variables are substituted consistently, function names and number of arguments match, and no cyclic substitutions (like X = f(X)) occur.
Valid: f(a, X) = f(a, b) → X = b
Invalid: X = f(X) (creates infinite loop)
🏷 Not for Examination – Only for Understanding Purpose
3. Define Unification Algorithm.
Answer:
The unification algorithm is a systematic procedure used to determine whether two logical expressions can be made identical by applying variable substitutions while resolving conflicts.
Step 1: Compare terms
Step 2: Substitute variable
Step 3: Check sub-terms
Step 4: Stop if conflict occurs
🏷 Not for Examination – Only for Understanding Purpose
4. State the conditions for successful unification.
Answer:
Successful unification requires matching constants, valid variable substitutions, matching function symbols with same arguments, and absence of cyclic dependencies.
color(X) = color(red) → X = red ✔
apple ≠ banana ✖
🏷 Not for Examination – Only for Understanding Purpose
5. Define Unification Algorithm.
Answer:
The unification algorithm is a rule-based method in AI that matches two predicate logic expressions by generating a consistent substitution set to make them identical.
loves(X, Y) = loves(Alice, Bob)
Substitution → X = Alice, Y = Bob
🏷 Not for Examination – Only for Understanding Purpose
6. List the components of a problem formulation in AI.
Answer:
The components of problem formulation include the initial state, goal state, actions (operators), state space, and path from initial to goal state.
Water Jug Problem:
Initial State → (0,0)
Goal State → (0,4)
Actions → Fill, Empty, Pour
🏷 Not for Examination – Only for Understanding Purpose
7. State about Partial-Order Planning (POP) with an example.
Answer:
Partial-Order Planning (POP) is a planning approach where actions are partially ordered and only necessary constraints are maintained, allowing flexibility in execution.
Wearing Shoes:
Left sock before left shoe
Right sock before right shoe
But left and right can be done in any order.
🏷 Not for Examination – Only for Understanding Purpose
8. Justify the need of Unification in AI.
Answer:
Unification is needed in AI for automated reasoning, theorem proving, and logic programming, as it enables pattern matching between facts and rules to derive new knowledge.
Rule: Parent(X, Y) → Child(Y, X)
Fact: Parent(John, Mary)
Unification helps infer → Child(Mary, John)
🏷 Not for Examination – Only for Understanding Purpose
9. Define State Space Search.
Answer:
State space search is a problem-solving method in AI where problems are represented as states and transitions, and a path is searched from the initial state to the goal state.
8-Puzzle Problem
Each board position = State
Moves of tiles = Actions
🏷 Not for Examination – Only for Understanding Purpose
10. List the components of State Representation.
Answer:
The components of state representation include state, initial state, goal state, transitions (actions), and path connecting the states.
Initial → (Start at Home)
Goal → (Reach College)
Actions → Walk, Bus, Bike
🏷 Not for Examination – Only for Understanding Purpose
11. Define Partial-Order Planning.
Answer:
Partial-Order Planning (POP) is a planning approach in AI where actions are partially ordered instead of strictly sequenced. Only necessary ordering constraints are imposed, allowing flexibility in execution.
Left sock must be worn before left shoe.
Right sock must be worn before right shoe.
But left and right sequences can happen in any order.
🏷 Not for Examination – Only for Understanding Purpose
12. Outline the role of Planning Graph in AI.
Answer:
A planning graph is a data structure used in AI planning that represents actions and states level by level. It helps in goal reachability analysis, heuristic computation, and detecting mutually exclusive actions.
Cake Problem:
State Level → Have(Cake)
Action Level → Eat(Cake)
Next State → Eaten(Cake)
🏷 Not for Examination – Only for Understanding Purpose
13. With an example brief about initial state and goal state in problem solving.
Answer:
The initial state is the starting configuration of a problem, and the goal state is the desired final configuration. Problem solving aims to find a path from the initial state to the goal state.
Water Jug Problem
Initial State → (0,0)
Goal State → (0,4)
🏷 Not for Examination – Only for Understanding Purpose
14. Mention any two limitations of planning with limited resources.
Answer:
Two limitations are: (1) High computational complexity due to resource constraints, and (2) Difficulty in optimizing time, cost, or energy simultaneously while achieving goals.
A robot may not complete all tasks due to low battery.
Delivery planning may fail due to limited fuel.
🏷 Not for Examination – Only for Understanding Purpose
15. Define Probability Theory in Artificial Intelligence.
Answer:
Probability theory in AI is a mathematical framework used to represent and manage uncertainty in knowledge and decision-making. It helps AI systems reason under uncertain conditions.
If P(Rain) = 0.7, there is a 70% chance of rain.
🏷 Not for Examination – Only for Understanding Purpose
16. Give two examples of uncertain knowledge.
Answer:
Two examples of uncertain knowledge are weather prediction and medical diagnosis, where outcomes cannot be determined with complete certainty.
“It may rain tomorrow.”
“A patient might have flu based on symptoms.”
🏷 Not for Examination – Only for Understanding Purpose
17. Express how planning is utilized in Forward March.
Answer:
In forward march (forward search), planning starts from the initial state and applies actions step by step until the goal state is reached. It expands successor states progressively.
Start → Fill Jug → Pour Water → Reach 4 Liters
🏷 Not for Examination – Only for Understanding Purpose
18. State any two uses of a Planning Graph.
Answer:
Two uses of a planning graph are: (1) Fast plan generation, and (2) Goal reachability analysis to determine if a goal can be achieved.
Used in robotics for quick task planning.
Used to check if cake can be eaten and baked again.
🏷 Not for Examination – Only for Understanding Purpose
19. List any two axioms of probability with example.
Answer:
Two axioms of probability are: (1) The probability of an event is between 0 and 1, and (2) The probability of the sample space is 1.
P(Head) = 0.5 (between 0 and 1)
P(Head or Tail) = 1
🏷 Not for Examination – Only for Understanding Purpose
20. Define Planning.
Answer:
Planning in AI is the process of selecting and organizing a sequence of actions to achieve a specific goal from a given initial state.
Goal → Reach College
Plan → Wake up → Take Bus → Arrive
🏷 Not for Examination – Only for Understanding Purpose
0 Comments