ARTIFICIAL INTELLIGENCE

LECTURE-4:
SOLVING PROBLEMS BY SEARCHING

By MUHAMMAD AFZAL
 Solving Problems by
2
Searching
 Sometimes, it's not easy for an agent to know
what to do next.
 When this happens, a agent might need to
think ahead.
 It can plan a series of steps to reach a goal. We
call this program a problem-solving agent, and
the process of planning is called searching.
 Search algorithms are distinguished as:
 informed algorithms: agent can estimate how far it
is from the goal
 uninformed algorithms: where no such estimate is
available.
 Problem-Solving Agents
3

 Problem-solving agent
 A kind of goal-based agent
 It solves problem by
 finding sequences of actions that lead to target
 If the environment is unknown and the agent has no
additional information, then it can execute one of
the action randomly
 Problem-Solving Agents
4

 To solve a problem, an agent can follow four-

stages process:
 1. GOAL FORMULATION
 2. PROBLEM FORMULATION
 3. SEARCH
 4. EXECUTION

 GOAL FORMULATION
 This is first stage of problem solving agent based on
the current situation.
 Goals help us focus on what we want to achieve.
 They limit what we try to do, so we can focus on
important things.
 Problem-Solving Agents
5

 PROBLEM FORMULATION: the agent plans a

description of the states and actions necessary to
reach the goal
 SEARCH: Before taking any action in the real
world, the agent simulates sequences of actions in
its model, searching until it finds a sequence of
actions that reaches the goal. Such a sequence is
called a solution
 EXECUTION: The agent can now execute the
actions in the solution, one at a time.
 Problem-Solving Agents
6

A simplified road map of part of Romania, with road distances in

 Problem-Solving Agents
7

 In a fully observable and known environment,

the solution to any problem is a fixed
sequence of actions: start from Arad drive to
Sibiu, then Fagaras, then Bucharest
 If the model is correct, and the agent has
found a solution, it can ignore its percepts
while it is executing the actions, This is called
an open-loop system.
 If the model is incorrect, or the environment
is unobservable, then the agent would be
safer using a closed-loop approach that
 Search Algorithms
8
 A search algorithm takes a search problem as
input and returns a solution, or an indication of
failure
 All problems are transformed as a search
tree.
 Search Tree: A tree representation of search
problem is called Search tree
 Search Space: Search space represents a set
of possible solutions, which a system may
have.
 Initial state

Three partial search
trees for finding a

Search Algorithms
route from Arad to
Bucharest

9
10
 Search Algorithms
11

 Now we must choose which of these three

child nodes to consider next.
 If we select Sibiu first, we call this the
frontier of the search tree
 Any state for which a node has been
generated is called as reached
 While a search tree may have cyclic paths:
A-B-A-B-A-B
 A good search strategy should avoid cyclic
paths
 Search Algorithms
12
 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 from initial state to the node

n itself
 DEPTH: number of steps along the path from the

initial state
 Measuring problem-solving
performance
13
 The evaluation of a search strategy
 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:
 howmuch memory is needed to perform the
search?
 Types of search algorithms
14

 Based on the search problems we can

classify the search algorithms into:
 Uninformed (Blind
search/Brute-force/exhaustive search)
algorithm
 An uninformed search algorithm is given no
clue about how close a state is to the goal(s).
 Informed search (Heuristic) search
algorithm
 An informed search knows which state is close
to the goal(s)
 Types of search algorithms
15
 Un-informed Search Strategies
16

 Breadth-first search (BFS):

 BFS algorithm starts searching from

the root node of the tree and expands

all successors node at the current level
before moving to nodes of the next
level then their successors, and so on.
 The breadth-first search algorithm is an

example of a general-graph search

algorithm
 Uninformed Search Strategies
17

 BFS uses an early goal test, checking

whether a node is a solution as soon as it is
generated

Breadth-first search on a simple

binary tree
 Un-informed Search Strategies
18

 Advantages:
 BFS will provide a solution if any solution exists
 If there are more than one solutions for a given
problem, then BFS will provide the minimal
solution which requires the least number of
steps
 Disadvantages:
 It requires lots of memory since each level of
the tree must be saved into memory to expand
the next level
 BFS needs lots of time if the solution is far away
from the root node
 Un-informed Search Strategies
19
 Example:
S---> A--->B---->C--->D---->G--->H--->E---->F---->I----
>K
 Un-informed Search Strategies
20

 Uniform Cost Search: This search

algorithm works when a different cost is
available for each edge
 It is a searching algorithm used for

traversing a weighted tree or graph

 It finds a path to the goal node which

has the lowest cumulative cost

 It can be used to solve any graph/tree

where the optimal cost is in demand

 Un-informed Search Strategies
21

The shortest path

here is
A−>B−>C−>E
forming the cost
17
 Un-informed Search Strategies
22

 Advantages:
 Uniform cost search is optimal because at every
state the path with the least cost is chosen
 Disadvantages:
 It does not care about the number of steps involve
in searching and only concerned about path cost
 Due to which this algorithm may be stuck in an
infinite loop
 Un-informed Search Strategies
23
 Depth-first search: Depth-first search always
expands the deepest node in the frontier first
 It starts from the root node and follows each path to
its greatest depth node before moving to the next
path
 Only when the search hits a dead end
 goes back and expands nodes at shallower levels
 Dead end  leaf nodes but not the goal
 DFS uses a stack data structure for its
implementation
Depth-first search
Depth-first search
 Depth-first search
26
 Advantage:
 DFS requires very less memory as it only needs

to store a stack of the nodes on the path from

root node to the current node
 It takes less time to reach to the goal node than

BFS algorithm (if it traverses in the right path)

 Disadvantage:
 DFS algorithm goes for deep down searching

and sometime it may go to the infinite loop

 Informed Search Strategies
27

 Informed Search also called as Heuristic

Search
 Informed search algorithms use domain
knowledge
 In an informed search, problem information is
available which can guide the search
 Informed search strategies can find a solution
more efficiently than an uninformed search
strategy
 Informed search can solve much complex
problem which could not be solved in another
 Informed Search Strategies
28

Best-first Search Algorithm (Greedy

Search):

 A greedy algorithm is an approach for

solving a problem by selecting the
best option available at the
moment
 Greedy best-first search algorithm

always selects the path which appears

best at that moment
 Best-first Search Algorithm (Greedy
Search):
29
 Best-first Search Algorithm (Greedy
Search):
30
 Best-first Search Algorithm (Greedy
Search):
31

 Advantages:
 This algorithm is more efficient than BFS and
DFS algorithms.
 Disadvantages:
 It can get stuck in a loop as DFS.
 This algorithm is not optimal.
 Informed Search Strategies
32

 A* search:
 A* search algorithm is a simple and efficient
that can be used to find the optimal path
between two nodes in a graph
 It searches for the shortest path between
the initial and the final state
 Informed Search Strategies
33
 Informed Search Strategies
34

 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.
 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.