Searching and Solving Problems

Introduction of Artificial Intelligence

 Simple reflex agents are limited in what they can do

 Because, their actions are determined only by the current percept

 Furthermore, they have no knowledge of what their actions do or of what they are trying to

achieve

 They can’t work well in environments

which this mapping would be too large to store and

would take too long to learn

 Hence, goal-based agent is used

2
 Introduction of Artificial Intelligence

 Goal-based agents can succeed by considering future actions and the

desirability of their outcomes

 Goal-based agents that use more advanced factored or structured

representations are usually called planning agents

 Uninformed and informed search algorithms are used to create

solution for problems

3
 Introduction of Artificial Intelligence

 Problem-solving agent is A kind of goal-based agent, It solves problem by finding sequences of

actions that lead to desirable states (goals)
 In Problem Solving Agent, there are four general steps:

Goal Formulation
 what are the successful world state
Problem Formulation
 what actions and states to consider give the goal

Search
 determine the possible sequence of actions that lead to the state of known values and the
choosing the best sequence
Execute
 Give the solution preform the actions 4
 Introduction of Artificial Intelligence

GOAL FORMULATION
 To solve a problem, the first step is the goal formulation, based on the current situation
 The goal is formulated as a set of world states, in which the goal is satisfied
 Goal-objectives that the agent is trying to achieve and, hence the actions it needs to consider.
 Reaching from initial state to à goal state
 Actions are required
 Actions are the operators
⚫ causing transitions between world states
⚫ Actions should be abstract enough at a certain degree, instead of very detailed

E.g., turn left VS turn left 30 degree, etc.

5
 Introduction of Artificial Intelligence

 Problem formulation: is the process of deciding what actions and states to consider, and

follows goal formulation

 An agent with several immediate options of unknown value can decide what to do by first

examining ;

 different possible sequences of actions that lead to states of known value, and then choosing

the best one

 This process of looking for such a sequence is called search

 A search algorithm takes a problem as input and returns a solution in the form of an action
6
sequence
 Introduction of Artificial Intelligence

 Once a solution is found, the actions it recommends can be carried out.

 This is called the execution phase

 Thus, we have a simple "formulate, search, execute" design for the agent

 There are four essentially different types of problems:

Single state problems,

Multiple-state problems,

Contingency problems, and

Exploration problems
7
 Introduction of Artificial Intelligence

 Well-defined problems and solutions

A problem is really a collection of information that the agent will use to decide what

to do

A problem is defined by 5 components:

Initial state

Actions

Transition model or (Successor functions)

 Goal Test

Path Cost 8
 Introduction of Artificial Intelligence

 State: Description of the state of the world in which the search agent finds itself.
 Starting / initial state: The initial state in which the search agent is started.
 Goal state: If the agent reaches a goal state, then it terminates and outputs a solution (if
desired).
 Actions: All of the agents allowed actions.
 Solution: The path in the search tree from the starting state to the goal state.
 Cost function: Assigns a cost value to every action. Necessary for finding a cost-optimal
solution.
 State space: Set of all states.
 A path in the state space- any sequence of states connected by a sequence of actions. 9
 Introduction of Artificial Intelligence

The initial state is that the agent knows itself to be in.

The set of possible actions available to the agent.

The term operator is used to denote the description of an action in terms of which state will

be reached by carrying out the action in a particular state

(An alternate formulation uses a successor function S.

Together, these define the state space of the problem: the set of all states reachable from the

initial state by any sequence of actions.

10
 Introduction of Artificial Intelligence

The goal test applied to the current state to test if the agent is in its goal

Sometimes there is an explicit set of possible goal states

Sometimes the goal is described by the properties instead of stating explicitly the set of

states

Example: Chess

o The agent wins if it can capture the KING of the opponent on next move

( checkmate).

o No matter what the opponent does

11
 Introduction of Artificial Intelligence

A path cost function is a function that assigns a cost to a path.

The cost of a path is the sum of the costs of the individual actions along the path.

The path cost function is often denoted by g

The solution of a problem is then a path from the initial state to a state satisfying the

goal test

to distinguish the best path from others

Optimal solution is the solution with lowest path cost among all solutions

12
 Introduction of Artificial Intelligence

 E.g:

13
 Introduction of Artificial Intelligence

1. Problem : To Go from Ajlun to Amman

2. Initial State : Ajlun
3. Operator / Action : Go from One City To another .
4. State Space : {Jarash , Salat , irbed, }
5. Goal Test : are the agent in Amman.
6. Path Cost Function : Get The Cost From The Map.

7. Solution :{ {Aj à Ja à Ir à Ma à Za à Am}, {ajàir àma àza àam}

……..{ajàja àam}}

14
 Introduction of Artificial Intelligence

 The 8-puzzle
Problem formulation
 State: the location of each of the eight tiles in one of the nine squares
 Operators: blank space moves left, right, up, or down
 Goal test: state matches the right figure (goal state)
 Path cost: each step costs 1, since we are moving one step each time.

15
 Introduction of Artificial Intelligence
EXERCISE:1 RIVER CROSSING PUZZLES
1. Goat, Wolf and Cabbage problem
 A farmer returns from the market, where he bought a goat a cabbage and a wolf. On the way home he must cross
a river. His boat is small and unable to transport more than one of his purchases. He cannot leave the goat alone
with the cabbage (because the goat would eat it), nor he can leave the goat alone with the wolf (because the goat

would be eaten). How can the farmer get everything safely on the other side?

1. Identify the set of possible states and operators

2. Construct the state space of the problem using suitable representation.
2.Towers of Hanoi puzzle
– The goal is to move all the discs from the left peg to the right one. Only one disc may be
moved at a time. A disc can be placed either on an empty peg or on top of a larger disc. Try
to move all the discs using the smallest number of moves possible.
16
 Introduction of Artificial Intelligence

Search Strategies

17
 Introduction of Artificial Intelligence

3. Search
 Because there are many ways to achieve the same goal
⚫ Those ways are together expressed as a tree

⚫ Multiple options of unknown value at a point,

 the agent can examine different possible sequences of actions, and choose the best
⚫ This process of looking for the best sequence is called search.

 The best sequence is then a list of actions, called solution

Search Algorithm:
 Defined as
⚫ taking a problem

⚫ and returns a solution

 Once a solution is found

⚫ the agent follows the solution

⚫ and carries out the list of actions – execution phase

 Design of an agent
 “Formulate, search, execute” 18
 Introduction of Artificial Intelligence

 SEARCHING FOR SOLUTIONS

 Search is the fundamental technique of AI.
 Possible answers, decisions or courses of action are structured into an abstract space, which is search.

 Finding out a solution is done by

 searching through the state space

 All problems are transformed

⚫ as a search tree

⚫ generated by the initial state and successor function

SEARCH TREE
 Initial state
⚫ The root of the search tree is a search node

 Expanding
⚫ applying successor function to the current state

⚫ There by generating a new set of states

 Leaf nodes
⚫ the states having no successors

19
 Introduction of Artificial Intelligence

 It is helpful to think of the search process as building up a search tree

 The root of the search tree is a search node corresponding to the initial state

 The leaf nodes of the tree correspond to states that do not have successors in the tree,

It is either because they have not been expanded yet, or

Because they were expanded, but generated the empty set

 At each step, the search algorithm chooses one leaf node to expand

20
 Introduction of Artificial Intelligence

 In search tree, a node is having five components:

⚫ STATE: which state it is in the state space
⚫ PARENT-NODE: from which node it is generated
⚫ ACTION: which action applied to its parent-node to generate it
⚫ PATH-COST: the cost, g(n), from initial state to the node n itself
⚫ DEPTH: number of steps along the path from the initial state

21
 Introduction of Artificial Intelligence

 Measuring problem-solving performance

Completeness: is the strategy guaranteed to find a solution when there is one?

Optimality: does the strategy find the highest-quality solution when there are several

different solutions?

Time complexity: how long does it take to find a solution?

Space complexity: how much memory is needed to perform the search?

22
 Introduction of Artificial Intelligence

In AI, complexity is expressed in

b, branching factor, maximum number of successors of any node

d, the depth of the shallowest goal node.

(depth of the least-cost solution)

m, the maximum length of any path in the state space

Time and Space is measured in

number of nodes generated during the search

maximum number of nodes stored in memory

23
Introduction of Artificial Intelligence

24
 Introduction of Artificial Intelligence

SEARCH METHODS:

 Uninformed search (blind search)

no information about the number of steps

or the path cost from the current state to the goal

search the state space blindly

 Informed search, or heuristic search

a cleverer strategy that searches toward the goal,

based on the information from the current state so far

25
 Introduction of Artificial Intelligence

 Uninformed search
⚫ Breadth first search
⚫ Depth first search
⚫ Uniform cost search
⚫ Depth limited search
⚫ Iterative deepening search

 Informed search
⚫ Greedy search
⚫ A*-search 26
 Introduction of Artificial Intelligence

 Breadth-first search

The root node is expanded first (FIFO)

All the nodes generated by the root node are then expanded

And then their successors and so on

In general, all the nodes at depth d in the search tree are expanded before the nodes

at depth d + 1

Can be implemented by GENERAL SEARCH algorithm

27
Introduction of Artificial Intelligence

28
Introduction of Artificial Intelligence

29
 Introduction of Artificial Intelligence

 EXAMPLE FOR ILLUSTRATING UNINFORMED SEARCH STRATEGIES

of Breadth first search

30
 Introduction of Artificial Intelligence

Breadth first search

Expanded node Nodes list

{ S0 }
S0 { A3 B1 C8 }
A3 { B1 C8 D6 E10 G18 }
B1 { C8 D6 E10 G18 G21 }
C8 { D6 E10 G18 G21 G13 }
D6 { E10 G18 G21 G13 }
E10 { G18 G21 G13 }
G18 { G21 G13 }
 Solution path found is S A G , cost 18
 Number of nodes expanded (including goal node) = 7
31
 Introduction of Artificial Intelligence

 Breadth first search analysis

Complete – find the solution eventually

Optimal, if step cost is 1

The disadvantage

if the branching factor of a node is large, for even small instances (e.g.,

chess)

the space complexity and the time complexity are enormous

32
 Introduction of Artificial Intelligence

 Properties of breadth-first search

Complete? Yes (if b is finite)

Time? 1+b+b2+b3+… +bd = b(bd-1) = O(bd+1)

Space? O(bd+1) (keeps every node in memory)

Optimal? Yes (if cost = 1 per step)

Space is the bigger problem (more than time)

33
 Introduction of Artificial Intelligence

Exercise 1
7
13

17
Goal State
4
1

Initial state

2
9

34
 Introduction of Artificial Intelligence

 Breadth-first finds the shallowest goal state

⚫ but not necessarily be the least-cost solution
⚫ work only if all step costs are equal
 Uniform cost search
⚫ modifies breadth-first strategy
o by always expanding the lowest-cost node
⚫ The lowest-cost node is measured by the path cost g(n)

35
 Introduction of Artificial Intelligence

 The first found solution is guaranteed to be the cheapest

⚫ least in depth

⚫ But restrict to non-decreasing path cost

 It is easy to see that breadth-first search is just uniform cost search with g(n) =

DEPTH(n)

 When certain conditions are met:

the first solution that is found is guaranteed to be the cheapest solution,

because if there were a cheaper path that was a solution,

it would have been expanded earlier, and thus would have been found first

36
 Introduction of Artificial Intelligence

 Consider the following route finding problem

A route-finding problem:
(a) The state space, showing the cost for
each operator,

(b) Progression of the search.

Each node is labelled with g(n).
At the next step, the goal node with g
= 10 will be selected

37
 Introduction of Artificial Intelligence

 EXAMPLE FOR ILLUSTRATING UNINFORMED SEARCH STRATEGIES

of uniform cost search

38
 Introduction of Artificial Intelligence

 Consider the above example in BFS

UNIFORM-COST SEARCH
Expanded node Nodes list
{ S0 }
S0………………………..{ B1 A3 C8 }
B1……………………….. { A3 C8 G21 }
A3…………………………{ D6 C8 E10 G18 G21 }
D6………………………… { C8 E10 G18 G1 }
C8……………………………{ E10 G13 G18 G21 }
E10………………………….{ G13 G18 G21 }
G13…………………………… { G18 G21 }
 Solution path found is S C G, cost =13
 Number of nodes expanded (including goal node) = 7

39
 Introduction of Artificial Intelligence

 Uniform Cost search (ANALYSIS)

 optimal
 complete
 Time and space complexities
L L
⚫ Reasonable, It takes O(b ) time and O(b ) space, where L is the depth limit

 suitable for the problem

⚫ having a large search space

⚫ and the depth of the solution is not known

 PROPERTIES OF Uniform depth SEARCH

¢ Complete? Yes
¢ Time? O(bL)
¢ Space? O(bL)
¢ Optimal? Yes, if step cost = 1
40
 Introduction of Artificial Intelligence

 Always expands one of the nodes at the

 deepest level of the tree

 Only when the search hits the end / Dead end

⚫ goes back and expands nodes at shallower levels
⚫ Dead end à leaf nodes but not the goal
 Backtracking search
⚫ only one successor is generated on expansion
⚫ rather than all successors
⚫ fewer memory
41
 Introduction of Artificial Intelligence

 Expand deepest unexpanded node Look A-N steps

 Implementations: A
……………………………….

……………………………………………………………………………………………
………………………………………………………………………..

N step

42
 Introduction of Artificial Intelligence

 Consider the above example in BFS

 DEPTH-FIRST SEARCH
Expanded node Nodes list
{ S0 }
S0 { A3 B1 C8 }
A3 { D6 E10 G18 B1 C8 }
D6 { E10 G18 B1 C8 }
E10 { G18 B1 C8 }
G18 { B1 C8 }

 Solution path found is S A G, cost 18

 Number of nodes expanded (including goal node) = 5

43
 Introduction of Artificial Intelligence

 Not complete
⚫ because a path may be infinite or looping
⚫ then the path will never fail and go back try another option
 Not optimal
⚫ it doesn't guarantee to the best solution
 It overcomes
• the time and space complexities

44
 Introduction of Artificial Intelligence

PROPERTIES OF DEPTH-FIRST SEARCH

 Complete? No: fails in infinite-depth spaces, spaces with loops

⚫ Modify to avoid repeated states along path à complete in finite spaces

 Time? O(bm) terrible if m is much larger than d

⚫ but if solutions are dense, may be much faster than breadth-first
 Space? O(bm), i.e., linear space!
 Optimal? No

45
 Introduction of Artificial Intelligence
ITERATIVE DEEPENING SEARCH
 Is a modified versions of DFS
 No choosing of the best depth limit
 It tries all possible depth limits:
⚫ first 0, then 1, 2, and so on

⚫ combines the benefits of depth-first and breadth-first

⚫ It will terminate when the depth limits reaches d, depth of shallowest goal

node, with success message.

46
 Introduction of Artificial Intelligence

Limit 1

47
 Introduction of Artificial Intelligence

 ITERATIVE DEEPENING SEARCH (ANALYSIS)

 Optimal: IDS algorithm is optimal if path cost is a non- decreasing function of the depth of
the node.
 Complete: this algorithm is complete is if the branching factor is finite.
 Time and space complexities
⚫ reasonable

 suitable for the problem

⚫ having a large search space

⚫ and the depth of the solution is not known

 PROPERTIES OF ITERATIVE DEEPENING SEARCH

¢ Complete? Yes
¢ Time? (d+1)b0 + d b1 + (d-1)b2 + … + bd = O(bd)
¢ Space? O(bd)
¢ Optimal? Yes, if step cost = 1 48
 Introduction of Artificial Intelligence

 Advantages:
It combines the benefits of BFS and DFS search algorithm in terms of fast search and
memory efficiency.
 Disadvantages:
The main drawback of IDS is that it repeats all the work of the previous phase.
Example:

I the third iteration, the algorithm will find the goal

49
 Introduction of Artificial Intelligence

 Depth-limited search avoids the pitfalls of depth-first search by imposing a cutoff on

the maximum depth of a path.

 This cutoff can be implemented with a special depth-limited search algorithm

 We are guaranteed to find the solution if it exists, but we are still not guaranteed to

find the shortest solution first.

 depth-limited search is Not complete and not optimal.

50
 Introduction of Artificial Intelligence

 If we choose a depth limit that is too small, then depth-limited search is not even complete.

 The time and space complexity of depth-limited search is similar to depth-first search

 It takes O(bl) time and O(bl) space, where l is the depth limit.

Example:

51
 Introduction of Artificial Intelligence

 Depth-limited search can be terminated with two Conditions of failure:

Standard failure value: It indicates that problem does not have any solution.

Cutoff failure value: It defines no solution for the problem within a given depth limit.

 Advantages:

Depth-limited search is Memory efficient.

 Disadvantages:

Depth-limited search also has a disadvantage of incompleteness.

It may not be optimal if the problem has more than one solution.
52
 Introduction of Artificial Intelligence

 Completeness: DLS search algorithm is complete if the solution is above the depth-

limit.

 Time Complexity: Time complexity of DLS algorithm is O(bℓ).

 Space Complexity: Space complexity of DLS algorithm is O(b×ℓ).

 Optimal: Depth-limited search can be viewed as a special case of DFS, and it is also

not optimal even if ℓ>d.

53
 Introduction of Artificial Intelligence

Run two simultaneous searches:

one forward from the initial state another backward from the goal

stop when the two searches meet

For problems where the branching factor is b in both directions:

 bidirectional search can make a big difference

54
 Introduction of Artificial Intelligence

If we assume as usual that there is a solution of depth d:

then the solution will be found in O(2bd/2) = O(bd/2) steps,

because the forward and backward searches each have to go only half way.

However, computing backward is difficult

The main question is, what does it mean to search backwards from the goal?

Searching backward means generating predecessors successively starting from the goal

node.

can the actions be reversible to compute its predecessors?

55
 Introduction of Artificial Intelligence

What can be done if there are many possible goal states?

We need to decide what kind of search is going to take place in each half

For example the following figure shows two breadth-first searches

56
 Introduction of Artificial Intelligence

 Advantages:
Bidirectional search is fast.
Bidirectional search requires less memory
 Disadvantages:
Implementation of the bidirectional search tree is difficult.
In bidirectional search, one should know the goal state in advance.

57
Introduction of Artificial Intelligence

58
 Introduction of Artificial Intelligence
INFORMED SEARCH/HEURISTIC SEARCH
 An informed search is more efficient than an uninformed search because in informed search,
along with the current state information, some additional information is also present, which make
it easy to reach the goal state.
Blind search – BFS, DFS, uniform cost
⚫ no notion concept of the “right direction”

⚫ can only recognize goal once it’s achieved

Heuristic search
 Estimates the cost of the minimal cost path from node n to a goal state
 h(n) >= 0 for all nodes n
 h(n) = 0 Implies that n is a goal node
 h(n) = ¥ Implies that n is a dead end from which a goal cannot be reached.
 All domain knowledge used in the search is encoded in the heuristic function h.
 Weak method :because of the limited way that domain-specific information is used to solve a
problem
 Sort nodes in the node list by increasing values of an evaluation function f(n) that incorporates
59
domain-specific information
 Introduction of Artificial Intelligence

 Idea: use an evaluation function f(n) for each node

⚫ estimate "desirability“(lack of it)

⚫ Expand most desirable unexpanded node

 When the nodes are ordered so that the one with the best evaluation expanded first, the
result is called best-first strategy

 A sample tree for best-first- search 60

 Introduction of Artificial Intelligence

Table: Search process for Best-First Search

Step # Node Children Available nodes that are Node chosen
being not yet expanded
expanded
1 S (A:3), (B:6), (C:5) (A:3), (C:5) (B:6) (A : 3)
2 A (D:9), (E:8) (B:6), (C:5), (D:9),(E : 8) (C:5)
3 C (H:7) (B:6), (D:9),(E : 8), H(:7) (B:6)
4 B (F:12), (G:14) (D:9), (E:8), (H:7), (F:12), (H:7)
(G:14)
5 H (I:5). (J:6) (D:9), (E:8), (F:12), (G:14), (I:5)
(I:5), (J:6)
6 I (K:1), (L:0), (M:2) (D:9), (E:8), (F:12), (G:14), (L,0)
(J:6), (K:1), (L:0), (M:2)
7 Search stops
61
 Introduction of Artificial Intelligence

A Best-first search that uses h to select the next node to expand is called Greedy Search

Goal: Minimizes the estimated cost to reach a goal

 expands the node that seems to be closest to a goal

 utilizes a heuristic function

 h(n) = estimated cost from the current node to a goal

 heuristic functions are problem-specific
 often straight-line distance(hSLD) for route-finding and similar problems
 often better than depth-first, although worst-time complexities are equal or
worse (space)
 Greedy search is neither optimal nor complete.
62
 Introduction of Artificial Intelligence

GREEDY SEARCH –EXAMPLE

 Use as an evaluation function f(n)=h(n), sorting nodes in the nodes list by increasing
values of f.
 Selects the node to expand that is believed to be closest
(i.e., smallest f values) to a goal node.
# of nodes tested 0, expanded 0

Expanded node Open list

(S:8)
63
 Introduction of Artificial Intelligence

GREEDY SEARCH –EXAMPLE

f(n)= h(n)
# of nodes tested 1, expanded 1

S
h=8

1 8

A B C
Expanded node OPEN list h=8 h=4 h=3

9 4

3 E 5
7
D G
(S:8) h= h= h=0

S not goal (C:3,B:4,A:8)

64
 Introduction of Artificial Intelligence

GREEDY SEARCH –EXAMPLE

f(n)= h(n)
# of nodes tested 2, expanded 2

S
h=8

Expanded node OPEN list 1 8

A B C
h=8 h=4 h=3

(S:8) 9 4

3 E 5
S
7
D G
(C:3,B:4,A:8) h= h= h=0

C is not goal (G:0,B:4,A:8)

65
 Introduction of Artificial Intelligence

GREEDY SEARCH –EXAMPLE

f(n)= h(n)
# of nodes tested 3, expanded 2
S
h=8

Expanded node OPEN list 1 8

A B C
h=8 h=4 h=3

(S:8) 9 4

3 E 5
7
S D G
(C:3,B:4,A:8) h= h= h=0

C (G:0,B:4,A:8)

Path: S,C,G Cost: 13

G goal * Fast but not (B:4.A:8)
optimal

66
 Introduction of Artificial Intelligence

PROPERTIES OF GREEDY BEST-FIRST SEARCH

 Complete?
⚫ No – can get stuck in loops,

 Time?
⚫ O(bm) – worst case (like Depth First Search)

⚫ But a good heuristic can give dramatic improvement

 Space?O(bm) – keeps all nodes in memory

 Optimal? No

67
 Introduction of Artificial Intelligence

 Advantages:
Best first search can switch between BFS and DFS by gaining the advantages of both
the algorithms.
This algorithm is more efficient than BFS and DFS algorithms.
 Disadvantages:
It can behave as an unguided depth-first search in the worst case scenario.
It can get stuck in a loop as DFS.
This algorithm is not optimal.

68
 Introduction of Artificial Intelligence

 Best-known form of best-first search.

 Goal: minimizes the total path cost
 Idea: avoid expanding paths that are already expensive.
 A∗ search uses an admissible heuristic
i.e., h(n) <= h∗(n) where h∗(n) is the true cost from n to a goal.
 (Also require h(n) = 0, so h(G) = 0 for any goal G.)
 E.g., h(n) never over estimates the actual road distance
 Theorem: A∗ search is optimal if h is admissible.
 Evaluation function f(n)=g(n) + h(n) in A*
⚫ g(n) the cost (so far) to reach the node.

⚫ h(n) estimated cost to get from the node to the goal.

⚫ f(n) estimated total cost of path through n to goal.

69
 Introduction of Artificial Intelligence
 Example for A* search
Initial state

Goal state

70
Introduction of Artificial Intelligence

71
 Introduction of Artificial Intelligence

 Advantages:
A* search algorithm is the best algorithm than other search algorithms.
A* search algorithm is optimal and complete.
This algorithm can solve very complex problems.
 Disadvantages:
It does not always produce the shortest path as it mostly based on heuristics and
approximation.
A* search algorithm has some complexity issues.
The main drawback of A* is memory requirement as it keeps all generated nodes in
the memory, so it is not practical for various large-scale problems.

72
 Introduction of Artificial Intelligence

Avoiding Repeated States

• Avoiding repeated states in search algorithms is critical to improving efficiency and
preventing unnecessary computations.
1. Use a Visited/Closed Set
• Maintain a set (or similar data structure) to track visited states. Before expanding a state:
• Check if the state is already in the visited set.
• If it is, skip processing that state. Otherwise, add the state to the visited set and proceed.
2. Use Recursive Backtracking with Marking
• In backtracking approaches, mark a state as visited before recursion and unmark it after
returning.
3. Iterative Deepening (for Depth-First Search)
• Limit the depth of exploration and incrementally increase it.
This can implicitly reduce the chance of revisiting irrelevant states in some scenarios.
73
 Introduction of Artificial Intelligence

Constraint Satisfaction problem

• A Constraint Satisfaction Problem (CSP) is a type of problem where the goal is to find
a solution that satisfies a set of constraints.
• CSPs are widely used in artificial intelligence (AI) for solving problems such as
scheduling, assignment, and resource allocation.
• We use a factored representation for each state: a set of variables, each of which has a
value.
• A problem is solved when each variable has a value that satisfies all the constraints on
the variable.
• A problem described this way is called a constraint satisfaction problem, or CSP.

74
 Introduction of Artificial Intelligence

Con’t…
• A constraint satisfaction problem consists of three components, Variable, Domain, and
Constraints.
• Variables:
The entities to be assigned values.
Example: X,Y,Z
• Domains:
The possible values that each variable can take.
Example: X∈{1,2,3}, Y∈{1,2}, Z ∈{a,b,c}
• Constraints:
Rules that restrict the values the variables can simultaneously take.
Example: X≠Y, X+Z≤5

75
 Introduction of Artificial Intelligence

Con’t…
• To solve a CSP, we need to define a state space and the notion of a solution.
• Each state in a CSP is defined by an assignment of values to some or all of the variables,
{Xi = vi, Xj = vj, . . .}.
• An assignment that does not violate any constraints is called a consistent or legal
assignment.
• A complete assignment is one in which every variable is assigned, and a solution to a
CSP is a consistent, complete assignment.
• A partial assignment is one that assigns values to only some of the variables.

76
 Introduction of Artificial Intelligence

Con’t…
Types of CSPs
• Finite CSP:
Variables have finite domains (e.g., scheduling, Sudoku).
• Infinite CSP:
Variables can take values from infinite domains (e.g., real numbers in equations).
• Binary CSP:
Constraints involve pairs of variables (e.g., X<Y).
• Higher-order CSP:
Constraints involve three or more variables (e.g., X+Y+Z=10, A<B<C)

77
 Introduction of Artificial Intelligence

Example of a CSP
• Problem:
• Imagine we have a map with four regions: A, B, C, and D. We want to color each
region such that no two adjacent regions have the same color. We have three colors
available: Red, Green, and Blue..
• Variables: Regions (e.g., A,B,C,D).
• Domains: Colors (e.g., {Red, Green, Blue}).
• Constraints: Adjacent regions cannot have the same color (e.g., A≠B, A≠C, B≠C,
B≠D,C≠D).

78
 Introduction of Artificial Intelligence

Solving CSPs
1. Backtracking Search:
• A systematic search algorithm that tries assigning values to variables and backtracks when
constraints are violated.
• Improvement Techniques:
• Forward Checking: Eliminate inconsistent values from domains of unassigned
variables.
• Constraint Propagation: Enforce constraints (e.g., Arc Consistency using AC-3
algorithm).
2. Local Search:
• Start with a random solution and iteratively improve by making small changes.
Example: Min-Conflicts heuristic for resolving violated constraints.

79
 Introduction of Artificial Intelligence

Con’t…
3. Heuristics:
• Most Constrained Variable (MRV): Assign a value to the variable with the fewest
legal values.
• Least Constraining Value (LCV): Choose the value that leaves the most options for
other variables.
4. Consistency Algorithms:
• Node Consistency: Ensure individual variables satisfy unary constraints.
• Arc Consistency (AC-3): Ensure every value of one variable has a consistent value in
another.

80
 Introduction of Artificial Intelligence

Applications of CSP
• Scheduling: Allocating resources to tasks subject to constraints.
• Timetabling: Assigning classes to timeslots without conflicts.
• Puzzle Solving: Sudoku, N-Queens problem.
• Configuration Problems: Designing systems subject to compatibility
constraints.
• Logical Reasoning: Modeling and solving logical inference problems.

81
 Introduction of Artificial Intelligence

Games as Search Problems

• Game theory views any multi-agent environment as a game, provided that the
impact of each agent on the others is “significant,” regardless of whether the agents
are cooperative or competitive.
• Games as Search Problems is a fascinating concept in artificial intelligence, where
solving a game is modeled as a search through a space of possible moves, leading to
a winning, losing, or drawn outcome.

82
 Introduction of Artificial Intelligence

Components of Games as Search Problems

• A game can be formally defined as a kind of search problem with the
following elements:
1. STATE SPACE: Represents all possible configurations of the game. For example, in
chess, each possible arrangement of pieces on the board is a state.
2. S0: The initial state, which specifies how the game is set up at the start
3. PLAYER(p): Defines which player has the move in a state.
4. ACTIONS(a): Returns the set of legal moves in a state.
5. RESULT(p, a): The transition model, which defines the result of a move
6. TERMINAL-TEST(s): A terminal test, which is true when the game is over and false
otherwise. States where the game has ended are called terminal states.
7. UTILITY(s, p): A utility function, defines the final numeric value for a game that ends
in terminal state s for a player p.
83
 Introduction of Artificial Intelligence

Example: Tic-Tac-Toe
• State Space: All possible board configurations (up to 39=19,683 states,
considering the 3x3 grid).
• Initial State: Empty board.
• Players: Player 1 (X) and Player 2 (O).
• Actions: Placing X or O in any empty cell.
• Transition Model: Updates the board after a move.
• Terminal States: A row, column, or diagonal is filled with the same symbol
(win), or no empty cells remain (draw).
• Utility Function:
• +1 for X wins, -1 for O wins, 0 for a draw.
84
 Introduction of Artificial Intelligence

Search Algorithms for Games

• Minimax Algorithm:
• A recursive strategy to minimize the possible loss in a worst-case scenario.
• Alternates between maximizing and minimizing players' moves.
• Alpha-Beta Pruning:
• Optimized version of Minimax.
• Prunes branches of the search tree that cannot affect the final decision.
• Monte Carlo Tree Search (MCTS):
• Uses random sampling to evaluate states, common in games with large state
spaces like Go.

85
 Introduction of Artificial Intelligence

Single-Agent vs. Multi-Agent Search

• Single-Agent Games: The search focuses on solving puzzles or finding optimal paths (e.g.,
Sudoku, Maze).
• Multi-Agent Games: Include adversarial dynamics, requiring strategies to handle
opponents (e.g., Chess, Checkers).

•Real-World Applications
• AI in Board Games: Chess engines like Stockfish, Go AI like AlphaGo.
• Video Games: AI controlling enemies or NPCs in real-time strategy or role-playing games.
• Decision-Making: Autonomous agents navigating adversarial environments.

86
 Introduction of Artificial Intelligence

THE ENDS OF CHAPTER 3

87