Scribd
Upload
161 views

AI - Problem Solving by Searching

The document discusses different approaches to problem solving, including the classical and intelligent agent approaches. The classical approach uses a generate-and-test method, trying solutions until a correct one is found. For the intelligent agent approach, problems are represented as graphs like state spaces or AND/OR graphs. Solving a problem then becomes a matter of searching the graph to find a path from the start to a goal state. Different search methods like depth-first, breadth-first, and heuristic searches are discussed for intelligently exploring the problem space.

Uploaded by

Adi Angel
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
161 views

AI - Problem Solving by Searching

The document discusses different approaches to problem solving, including the classical and intelligent agent approaches. The classical approach uses a generate-and-test method, trying solutions until a correct one is found. For the intelligent agent approach, problems are represented as graphs like state spaces or AND/OR graphs. Solving a problem then becomes a matter of searching the graph to find a path from the start to a goal state. Different search methods like depth-first, breadth-first, and heuristic searches are discussed for intelligently exploring the problem space.

Uploaded by

Adi Angel
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 28

PROBLEM SOLVING AND SEARCH

Classical Approach
 The classical approach takes hit and trial
method to have solution of a problem.

 The intelligent species use this approach to


solve problems at hand.

 It is also called a generate and test approach.

 There is a process of solution generation.


Classical Approach
 Then the solution is verified.

 If found correct it is adopted and we finish.

 Otherwise we discard the solution and pick a


new solution to verify.

 This process iterates until we pick a correct


solution.
Representing a problem for solution.
 Problems generally represented as graphs

 Problem solving ~ searching a graph

 Two representations

(1) State space (usual graph)


(2) AND/OR graph
What is State Space?
 State space = Directed graph
 Nodes ~ Problem situations
 Arcs ~ Actions, legal moves

 Problem = ( State space, Start, Goal condition)


 Note: several nodes may satisfy goal condition

 Solving a problem ~ Finding a path


 Problem solving ~ Graph search
 Problem solution ~ Path from start to a goal node
Intelligent Agent’s approach
 Intelligent agents are supposed to maximize the
performance measure.

 This can be achieved by adopting a goal.

 Goals help organize behavior by limiting the


objectives that the agent is trying to achieve.
 Goal formulation is based on the current
situation and the agent’s performance measure
(It is the first step in problem solving).
Intelligent Agent’s approach
 Goal could be a set of states.

 The agent’s task is to find out which sequence


of actions will get it to a goal state.

 Problem representation is the process of


deciding what sorts of actions and states to
consider, given a goal.
Intelligent Agent’s approach
 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 a states of known value,
and then choosing the best sequence.

 Looking for such a sequence is called Search.

 A searching algorithm takes a problem as input


and returns solution in the form of action
sequence.
A problem from blocks world

Find a sequence of robot moves to re-arrange blocks


Blocks world state space
Start

Goal
Examples of representing problems
in state space
 Blocks world planning
 8-puzzle, 15-puzzle
 8 queens
 Travelling salesman
 Set covering

How can these problems be represented by graphs?

Propose corresponding state spaces


8-puzzle
State spaces for optimisation
problems
 Optimisation: minimise cost of solution

 In blocks world:
actions may have different costs
(blocks have different weights, ...)

 Assign costs to arcs


 Cost of solution = cost of solution path
More complex examples
 Making a time table
 Production scheduling
 Grammatical parsing
 Interpretation of sensory data
 Modelling from measured data
 Finding scientific theories that account for
experimental data
SEARCH METHODS
 Uninformed techniques:
systematically search complete graph,
unguided
 Informed methods:
Use problem specific information to guide
search in promising directions
 What is “promising”?
 Domain specific knowledge
 Heuristics
Basic search methods - uninformed

 Depth-first search
 Breadth-first search
 Iterative deepening
Informed, heuristic search

 Best-first search
 Hill climbing, steepest descent
 Algorithm A*
 Beam search
 Algorithm IDA* (Iterative Deepening A*)
 Algorithm RBFS (Recursive Best First
Search)
Direction of search
 Forward search: from start to goal

 Backward search: from goal to start

 Bidirectional search

 In expert systems:
Forward chaining
Backward chaining
Depth-first search
Properties of depth-first search
program
 Is not guaranteed to find shortest solution first
 Susceptible to infinite loops (should check for
cycles)
 Has low space complexity: only proportional to
depth of search
 Only requires memory to store the current path
from start to the current node
 When moving to alternative path, previously
searched paths can be forgotten
Iterative deepening search
 Depth-limited search may miss a solution if
depth-limit is set too low

 This may be problematic if solution length not


known in advance

 Idea: start with small MaxDepth and increase


MaxDepth until solution found
Breadth-first search

•Guaranteed to find shortest solution first


•best-first finds solution a-c-f
•depth-first finds a-b-e-j
A breadth-first search program
 Breadth-first search completes one level
before moving on to next level
 Has to keep in memory all the competing paths
that aspire to be extended to a goal node
 A possible representation of candidate paths:
list of lists
 Easiest to store paths in reverse order;
then, to extend a path, add a node as new
head (easier than adding a node at end of list)
Candidate paths as list of lists

a
b c
d e f g

[[[c,a],
[d,b,a],
[[ [a] ][d,b,a],
[b,a], initial
[e,b,a],
[c,a] list[f,c,a],
][e,b,a]of candidate
after [g,c,a]
] expanding
after ] paths
a
expanding b

On each iteration: Remove first candidate path,


extend it and add extensions at end of list
Complexity of basic search methods

For simpler analysis consider state-space as a tree


Uniform branching b
Solution at depth d

Number of nodes at level n : bn


Time and space complexity orders

Shortest
solution
Time Space guaranteed

Breadth-first bd bd yes
Depth-first bdmax dmax no
Iterative deepening bd d yes
That’s All
Thank You
28

You might also like