Transcript Search Problems Russell and Norvig: Chapter 3, Sections 3.1 – 3.3

Search Problems
Russell and Norvig:
Chapter 3, Sections 3.1 – 3.3
Problem-Solving Agent
sensors
?
environment
agent
actuators
Search Problems
2
Problem-Solving Agent
sensors
?
environment
agent
actuators
• Actions
• Initial state
• Goal test
Search Problems
3
State Space and Successor Function
state space
successor function
• Actions
• Initial state
• Goal test
Search Problems
4
Initial State
state space
successor function
• Actions
• Initial state
• Goal test
Search Problems
5
Goal Test
state space
successor function
• Actions
• Initial state
• Goal test
Search Problems
6
Example: 8-puzzle
8
2
3
4
5
1
1
2
3
7
4
5
6
6
7
8
Initial state
Goal state
Search Problems
7
Example: 8-puzzle
8
2
3
4
7
5
1
6
8
2
3
4
5
1
8
7
6
2
8
2
3
4
7
3
4
7
5
1
6
5
1
6
Search Problems
8
Example: 8-puzzle
Size of the state space = 9!/2 = 181,440
15-puzzle  .65 x 1012
0.18 sec
6 days
24-puzzle  .5 x 1025
12 billion years
10 millions states/sec
Search Problems
9
Search Problem
State space
Initial state
Successor function
Goal test
Path cost
Search Problems
10
Search Problem
State space


each state is an abstract representation of
the environment
the state space is discrete
Initial state
Successor function
Goal test
Path cost
Search Problems
11
Search Problem
State space
Initial state:


usually the current state
sometimes one or several hypothetical
states (“what if …”)
Successor function
Goal test
Path cost
Search Problems
12
Search Problem
State space
Initial state
Successor function:


[state  subset of states]
an abstract representation of the possible
actions
Goal test
Path cost
Search Problems
13
Search Problem
State space
Initial state
Successor function
Goal test:


usually a condition
sometimes the description of a state
Path cost
Search Problems
14
Search Problem
State space
Initial state
Successor function
Goal test
Path cost:



[path  positive number]
usually, path cost = sum of step costs
e.g., number of moves of the empty tile
Search Problems
15
Search of State Space
Search Problems
16
Search of State Space
Search Problems
17
Search State Space
Search Problems
18
Search of State Space
Search Problems
19
Search of State Space
Search Problems
20
Search of State Space
 search tree
Search Problems
21
Simple Agent Algorithm
Problem-Solving-Agent
1. initial-state  sense/read state
2. goal  select/read goal
3. successor  select/read action models
4. problem  (initial-state, goal, successor)
5. solution  search(problem)
6. perform(solution)
Search Problems
22
Example: 8-queens
Place 8 queens in a chessboard so that no two queens
are in the same row, column, or diagonal.
A solution
Not a solution
Search Problems
23
Example: 8-queens
Formulation #1:
• States: any arrangement of
0 to 8 queens on the board
• Initial state: 0 queens on the
board
• Successor function: add a
queen in any square
• Goal test: 8 queens on the
board, none attacked
 648 states with 8 queens
Search Problems
24
Example: 8-queens
 2,067 states
Formulation #2:
• States: any arrangement of
k = 0 to 8 queens in the k
leftmost columns with none
attacked
• Initial state: 0 queens on the
board
• Successor function: add a
queen to any square in the
leftmost empty column such
that it is not attacked
by any other queen
• Goal test: 8 queens on the
Search
Problems
board
25
실제 n-queen 문제
Neural, Genetic 또는 Heuristic 방법으로 잘
해결


최악의 경우에는 처리 불가능
실제 n이 커지면 답이 매우 많으므로 간단한
Heuristics로도 답을 쉽게 찾음
 따라서 n이 커도 답을 잘 찾는다고 해서 인공지능
접근방법이 문제를 해결한다는 증거는 아님
 그러나 많은 실제 문제는 알고리즘에서 이야기하는
최악의 경우로는 잘 가지 않음
 더구나 대부분 우리가 원하는 답은 최적이 아니라 실제
활용해서 도움이 되는, feasible solution을 원하므로
인공지능 기법이 효과적으로 이용될 수 있음
Search Problems
26
Example: Robot navigation
What is the state space?
Search Problems
27
Example: Robot navigation
Cost of one horizontal/vertical step = 1
Cost of one diagonal step = 2
Search Problems
28
Example: Robot navigation
Search Problems
29
Example: Robot navigation
Search Problems
30
Example: Robot navigation
Cost of one step = ???
Search Problems
31
Example: Robot navigation
Search Problems
32
Example: Robot navigation
Search Problems
33
Example: Robot navigation
Cost of one step: length of segment
Search Problems
34
Example: Robot navigation
Search Problems
35
Example: Assembly Planning
Initial state
Complex function:
it must find if a collision-free
merging motion exists
Goal state
Successor function:
• Merge two subassemblies
Search Problems
36
Example: Assembly Planning
Search Problems
37
Example: Assembly Planning
Search Problems
38
Assumptions in Basic Search
The
The
The
The
environment is static
environment is discretizable
environment is observable
actions are deterministic
 open-loop solution
Search Problems
39
Search Problem Formulation
Real-world environment  Abstraction
Search Problems
40
Search Problem Formulation
Real-world environment  Abstraction

Validity:
 Can the solution be executed?
Search Problems
41
Search Problem Formulation
Real-world environment  Abstraction

Validity:
 Can the solution be executed?
 Does the state space contain the solution?
Search Problems
42
Search Problems
43
Search Problems
44
Search Problems
45
Search Problems
46
Search Problem Formulation
Real-world environment  Abstraction

Validity:
 Can the solution be executed?
 Does the state space contain the solution?

Usefulness
 Is the abstract problem easier than the real-
world problem?
Search Problems
47
Search Problem Formulation
Real-world environment  Abstraction

Validity:
 Can the solution be executed?
 Does the state space contain the solution?

Usefulness
 Is the abstract problem easier than the real-
world problem?
Without abstraction an agent would be
swamped by the real world
Search Problems
48
Search Problem Variants
One or several initial states
One or several goal states
The solution is the path or a goal node


In the 8-puzzle problem, it is the path to a
goal node
In the 8-queen problem, it is a goal node
Search Problems
49
Problem Variants
One or several initial states
One or several goal states
The solution is the path or a goal node
Any, or the best, or all solutions
Search Problems
50
Important Parameters
Number of states in state space
8-puzzle  181,440
15-puzzle  .65 x 1012
24-puzzle  .5 x 1025
8-queens  2,057
100-queens  1052
There exist techniques to solve
N-queens problems efficiently!
Stating a problem as a search problem
is not always a good idea!
Search Problems
51
Important Parameters
Number of states in state space
Size of memory needed to store a state
Search Problems
52
Important Parameters
Number of states in state space
Size of memory needed to store a state
Running time of the successor function
Search Problems
53
Applications
Route finding: airline travel,
telephone/computer networks
Pipe routing, VLSI routing
Pharmaceutical drug design
Robot motion planning
Video games
Search Problems
54
Task Environment
Observable
Static
Discrete
Agents
Crossword puzzle
Fully
Static
Discrete
Single
Chess with a clock
Fully
Strategic
Sequential
Semi
Discrete
Multi
Partially
Strategic
Sequential
Static
Discrete
Multi
Fully
Stochastic
Sequential
Static
Discrete
Multi
Taxi driving
Partially
Stochastic
Sequential
Dynamic
Continuous
Multi
Medical diagnosis
Partially
Stochastic
Sequential
Dynamic
Continuous
Single
Fully
Deterministic
Episodic
Semi
Continuous
Single
Part-picking robot
Partially
Stochastic
Episodic
Dynamic
Continuous
Single
Refinery controller
Partially
Stochastic
Sequential
Dynamic
Continuous
Single
Interactive English tutor
Partially
Stochastic
Sequential
Dynamic
Discrete
Multi
Poker
Backgammon
Image-analysis
Figure 2.6
Deterministic
Episodic
Deterministic Sequential
Examples of task environments and their characteristics.
Search Problems
55
Summary
Problem-solving agent
State space, successor function, search
Examples: 8-puzzle, 8-queens, route
finding, robot navigation, assembly
planning
Assumptions of basic search
Important parameters
Search Problems
56
Future Classes
Search strategies


Blind strategies
Heuristic strategies
Extensions



Uncertainty in state sensing
Uncertainty action model
On-line problem solving
Search Problems
57