MIRPUR UNIVERSITY OF SCIENCE AND TECHNOLOGY

DEPARTMENT OF SOFTWARE ENGINEERING

Artificial Intelligence
(SOLVING PROBLEMS BY
SEARCHING)

(lecture # 7)

Engr. Abdul Qadir Khan

(Lecturer)
 LECTURE CONTENTS

• Heuristic in AI
• No of Missing Tiles, Manhattan Distance
• Best First Search
• Greedy Best Search

Artificial Intelligence 3
 Heuristic in AI

• What- Heuristics are problem-solving strategies or rules of

thumb that simplify decision-making processes in AI.
• WHY-Heuristics are employed to find quick and satisfactory
solutions in situations where exhaustive search or
computation is impractical.
Efficiency:
• They are essential for making AI algorithms more efficient,
enabling them to handle complex problems in a reasonable
amount of time.

Artificial Intelligence 4
 Heuristic in AI

Problem-Solving:
Heuristics are commonly used in problem-solving tasks,
allowing AI systems to explore solution spaces more
effectively.
The trade-off with Accuracy:
While heuristics provide quick solutions, they may not
guarantee the most accurate result. There's often a trade-off
between speed and accuracy.
- good solution but not optimal always.

Artificial Intelligence 5
 How? Euclidean distance

Euclidean distance is a measure of the straight-line distance between two points in Euclidean
space. In two-dimensional space, it is the length of the shortest path between two points, forming
a right-angled triangle. The formula for calculating Euclidean distance between two points
(x1,y1) and (x2​,y2​) in a two-dimensional space is given by:
Euclidean Distance = √

In three-dimensional space, the formula extends to:

= √

And in n-dimensional space, the general formula is:

Artificial Intelligence 6
 How?

Artificial Intelligence 7
 Example: 8-puzzle

• Heuristic : Number of Missing Tiles

1 2 3 1 2 3

4 6 5 4 5 6

8 7 7 8

Initial state Goal State

March 19, 2024
Artificial Intelligence 8
 Example: 8-puzzle
• Heuristic: no of Misplaced tiles

1 2 3
4 6 5 no ono of Misplaced tiles=4
8 7 f Misplaced tiles=4

1 2 3 1 2 3
4 6 4 6 5
no of Mino of Misplaced tiles=4 nMino of Misplaced tiles=4
8 7 5 8 7 o of Misplaced tiles=4
splaced tiles=4

Artificial Intelligence 9
 Example: 8-puzzle
• Heuristic: no of Misplaced tiles

1 2 3
4 6 5 no of Misplaced tiles=4
8 7

1 2 3 1 2 3
4 6 4 6 5
no of Misplaced tiles=4 no of Misplaced tiles=4
8 7 5 8 7

1 2 3 1 2 3
no of Misplaced tiles=3 4 6 4 6
8 7 5 8 7 5
Artificial Intelligence 10
 How? Manhattan distance

Heuristic: Manhattan distance of tiles from the

correct location

Artificial Intelligence 11
 Example: 8-puzzle

1 2 3 1 2 3

4 6 5 4 5 6

8 7 7 8

Initial state Goal State

Artificial Intelligence 12
 Example: 8-puzzle

1 2 3
4 6 5 distance = 0+0+0+0+1+1+1+1=4
8 7

1 2 3 1 2 3
4 6 4 6 5
distance = 0+0+0+0+1+1+1+2=5 distance = 0+0+0+0+1+1+1+2=5
8 7 5 8 7

March 19, 2024

Artificial Intelligence 13
 Best First Search

• Best-First Search is a generic search algorithm that selects the most

promising node for expansion based on a heuristic evaluation.
• Best-First Search relies on a heuristic function to evaluate nodes in
the search space. The heuristic estimates the cost or desirability of
reaching the goal from a particular node
• The algorithm selects the node with the best heuristic value for
expansion.
• A priority queue is typically used to maintain the open set of nodes.
Nodes are ordered in the priority queue based on their heuristic
values.

Artificial Intelligence 14
 Best First Search

• Best-First Search often follows a greedy approach, prioritizing

immediate gains based on the heuristic information without
considering the overall path cost.
• While Best-First Search can be efficient, especially with well-designed
heuristics, it does not guarantee optimality.
• The algorithm's completeness depends on the nature of the problem
and the heuristic used.

Artificial Intelligence 15
 Algorithm
Algorithm: Best-First-Search(Graph g, Node start)
1. Create an empty PriorityQueue
PriorityQueue pq;
2. Insert "start" in pq.
pq.insert(start);
3. Until PriorityQueue is empty
u = pq.deleteMin;
If u is the goal
Exit
Else
For each neighbor v of u
If v is "Unvisited"
Mark v "Visited"
pq.insert(v)
Mark u "Examined"
End Algorithm
Artificial Intelligence 16
 Romania - Step Costs in KM

Artificial Intelligence 17
 Greedy Best-First Search

• Evaluation function h(n) (heuristic)

• Estimated cost of the cheapest path from n to a goal node
• E.g., hSLD(n) = straight-line distance from n to Bucharest
• Greedy search expands the node that appears to be closest to
goal.

March 19, 2024

Artificial Intelligence 18
 Greedy Best-First search example

Artificial Intelligence
 Greedy Best-First search example

Artificial Intelligence
 Greedy Best-First search example

Artificial Intelligence
 Greedy Best-First search example

March 19, 2024

Artificial Intelligence 22
 Properties of Greedy Best-First search

• Complete? No – can get stuck in loops, e.g., with

Oradea as goal and start from Iasi:
• Iasi  Neamt  Iasi  Neamt 
• Complete in finite space with repeated state checking

• Time? O(bm), but a good heuristic can give dramatic

improvement

• Space? O(bm) -- keeps all nodes in memory

• Optimal? No.
Artificial Intelligence
THANKS