Transcript 323-670 ปัญญาประดิษฐ์ (Artificial Intelligence)

344-471
ปัญญาประดิษฐ์และระบบชานาญการ
(Artificial Intelligence & Expert Systems)
ดร.วิภาดา เวทย์ประสิ ทธิ์
ภาควิชาวิทยาการคอมพิวเตอร์ คณะวิทยาศาสตร์ มหาวิทยาลัยสงขลานคริ นทร์
ห้องทางาน : M100/3 โทรศัพท์ : 074-288702
E-mail : [email protected]
Web site : http:// www.cs.psu.ac.th/wiphada
1
วัตถุประสงค์
1. ให้นกั ศึกษามีความรู ้ความเข้าใจเกี่ยวกับปั ญญาประดิษฐ์และสาขาต่างๆของ
ปัญญาประดิษฐ์
2. ให้นกั ศึกษาสามารถพัฒนางานทางด้านปัญญาประดิษฐ์ได้
3. ให้นกั ศึกษาสามารถค้นคว้าเพิ่มเติมด้วยตนเองได้
วิธีการเรียนการสอน : การบรรยาย การสัมมนา การศึกษาค้นคว้าด้วยตัวเอง
การวัดผล :
สอบกลางภาค 30%
สอบปลายภาค 30%
สัมมนา/การบ้าน 10% Assignment 20%
เวลาเรียน :
จันทร์ 10 - 11 ห้อง วท1
พุธ 9 –10 ห้อง M.309
พฤหัส 9 –10 ห้อง M.309 และ 10-11 ห้อง L5
2
344-471 AI & ES
Chapter 1
เนื้อหาวิชา
เนื้อหา
Chapter 1 : What is Artificial Intelligence?
Chapter 2 : Heuristic Search techniques
Chapter 3 : Knowledge Representation
Chapter 4 : Predicate Logics
Chapter 5 : Expert Systems
Chapter 6 : Natural Language Processing
Chapter 7 : Robotics
Chapter 8 : Computer vision
Chapter 9 : Neural Networks
Chapter 10 : Machine Learning
344-471 AI & ES
3
งาน
Homework 1
Homework 2
Homework 3
Lab 1 : สัมมนา
Lab 2 : สัมมนา
Lab 3 : สัมมนา
Lab 4 : สัมมนา
Lab 5 : สัมมนา
Lab 6 : สัมมนา
Chapter 1
เอกสารอ้างอิง
1. Artificial Intelligence second edition, Elaine Rich and Kevin Knight,
McGraw-Hill Inc., 1991.
2. Neural Networks : A Comprehensive Foundation, Simon Haykin, Macmillan
College Publishing Company Inc. 1994.
2. Artificial Intelligence : Structures and Strategies for Complex Problem
Solving, Second Edition, 1993.
3. Expert Systems and Applied Artificial Intelligence, Efraim Turban,
Macmillan Publishing Company, 1992.
4. Understanding Decision Support Systems and Expert Systems, Efrem G.
Mallach, Richard D. Irwin, Inc. 1994
5. Introduction to Natural Language Processing, Mary Dee Harries, 1985.
4
344-471 AI & ES
Chapter 1
Chapter 1
What is Artificial Intelligence?
5
AI Areas
Artificial Intelligence (AI) :
the branch o f computer science that is concerned with the
automation of intelligent behavior.
AI Areas :
Game Playing
 Automated Reasoning and Theorem Proving
 Expert Systems
 Natural Language Understanding and Semantics Modeling
 Modeling Human Performance
 Planning and Robotics
 Machine Leaning
 Neural Networks

6
344-471 AI & ES
Chapter 1
Task Domain of AI
Mundane Tasks
Perception :
Vision, Speech
Natural language :Understanding, Generation, Translation
Commonsense reasoning
Robot control
Formal Tasks
Games:
Chess
Mathematics :
Logic, Geometry






Expert Tasks
Engineering :
Design, Fault finding, Manufacturing planning
Scientific analysis
Medical diagnosis
Financial analysis




7
344-471 AI & ES
Chapter 1
Intelligence require knowledge
1.
2.
3.
4.
8
It is voluminous.
Is it shard to characterize accurately.
It is constantly changing.
It differs from data by being organized in
a way that corresponds to the ways it
will be used.
344-471 AI & ES
Chapter 1
Knowledge Representation and Search for AI
1.
2.
3.
4.
5.
9
The knowledge captures generalizations.
It can be understood by people who must provide it.
It can easily be modified to correct errors and to
reflect changes in the world.
It can be used in many situations even if it is not
totally accurate or complete.
It can use to narrow the range of possibilities that
must usually be considered.
344-471 AI & ES
Chapter 1
Common Features of AI Problems
1.
2.
3.
4.
5.
6.
7.
10
The use of computer to do the symbolic reasoning.
A focus on problems that do not respond to algorithmic
solutions.  Heuristic search.
Manipulate the significant quantitative features of a
situation rather than relying on numeric methods.
Dealing with semantic meaning.
Answer that are neither exact nor optimal but
“sufficient”.
Domain specific knowledge in solving problems.
Use meta-level knowledge.
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 1
11
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 1
12
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 1
1
2
3
4
5
6
7
8
9

Board : nine element vector representation.
0 = blank,
1 =X,
2=O

Moveable : Their Complexity = 39 = 19,683

 view vector board as a ternary number (base three)
13
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 2
14
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 2
15
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 2


Board : nine element vector representation.
2 = blank,
3 =X,
5=O






16
2
3
4
5
6
7
8
9
an integer indicating which move of the game is about to played.
1 indicate the first move.
9 indicate the last move.
Board[5] = 2  mean blank
Poswin(p) : If it produce (3*3*2) =18  X can win


1
p = 0 if the player can not win on his next move.
Poswin(p) : If it produce (5*5*2) =50 O can win
Go(n) : Make a move on square n.
 TURN is odd  if it is playing X
 TURN is even  if it is playing O
 More efficient in term of space.
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 2’
17
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 2’
18
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 2’


Board : nine element vector representation.
2 = blank,
3 =X,
5=O





19
3
4
1
5
9
6
7
2
an integer indicating which move of the game is about to played.
1 indicate the first move.
9 indicate the last move.
Board[5] = 2  mean blank
Poswin(p) : If it produce MAGIC SQUARE
 (8 + 3 + 4) =15


8
p = 0 if the player can not win on his next move.
Go(n) : Make a move on square n.
 TURN is odd  if it is playing X
 TURN is even  if it is playing O
 More efficient in term of space.
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 3
20
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 3
21
344-471 AI & ES
Chapter 1
Tic-Tac-Toe : Program 3





2
3
4
5
6
7
8
9
Board_Position : nine element vector representing the board, a
list of board positions that could result from the next move,
and a number representing as estimate of how likely the board
position is lead to an ultimate win for the player to move.
Minimax Procedure : in chapter 12.
Search tree : need more time
Use AI technique :
Decide which of a set of board positions is best.



22
1
find highest possible rating.
consider all the moves the component could make next.
 See which move is worst for us....
(Assume the opponent will make that move)
344-471 AI & ES
Chapter 1
Question Answering 1
Russia massed troops on the Czech border.
POLITICS program [Corbonell,1980)
Q1: Why did Russia do this?
A1: รัสเซียต้องการเข้าแทรกแซงการเมืองในเชคโกสโลวาเกีย.....
Q1: What should the United States do?
A2: ส่ งทหารไปช่ วยรบกับเชคโกสโลวาเกีย OR
A2:ประณามการกระทาของรัสเซีย
23
344-471 AI & ES
Chapter 1
Question Answering 2
Mary
went shopping for a new coat.
She found a red one she really liked.
When she got it home, she discovered that it went perfectly
with her favorite dress.
ELIZA
Q1:What did Mary go shopping for?
A1: .............................................
Q2:What did Mary find she liked?
A2:.............................................
Q3: Did Mary buy anything ?
A3:.............................................
24
344-471 AI & ES
Chapter 1
The level of the model
1.
2.
3.
25
What is the goal in trying to produce
programs that do intelligent things that people
do?
Are we trying to produce programs that do the
tasks the same way people do?
Are we attempting to produce programs that
simply do the tasks in whatever way appears
easiest?
344-471 AI & ES
Chapter 1
Model human performance
1. To test psychological theories of human
performance.
วิเคราะห์นิสยั เกเรของคนไข้ PAPPY {Colby, 1975]
2. To enable computers to understand human
reasoning.
อ่านหนังสื อพิมพ์และวิเคราะห์ข่าวได้
3. To enable computers to understand computer
reasoning.
 เข้าใจว่า คอมพิวเตอร์ประมวลผลลัพธ์ต่างๆมาได้อย่างไร
26
344-471 AI & ES
Chapter 1
Criteria for success

How will we know if we have succeeded?

Turing test.

DENDRAL : is a program that analyzes organic
compounds to determine their structure.
Human
Computer Person asking?
HUMAN CHEMIST
27
COMPUTER
344-471 AI & ES
Chapter 1
Homework 1 (2%)
1.
2.
Given the meaning of Artificial Intelligence
from your point of view. You may add citation
from searching documents in the web or from
the text book.
Given all AI fields with some explanations.
Due date : Wednesday 17 November 2004 in class
28
344-471 AI & ES
Chapter 1