Chapter 1 Sc

7/27/2019 Chapter 1 Sc

http://slidepdf.com/reader/full/chapter-1-sc 1/12

8/28/200

Page 1

Soft Computing

1

Neuro-Fuzzy and Soft Computing

chapter 1

J.-S.R. Jang

Neuro-Fuzzy and Soft Computing

chapter 1

J.-S.R. Jang

Bill CheethamBill Cheetham

Kai GoebelKai Goebel

Soft Computing

2

What is covered in this class?What is covered in this class?

We will teach techniques useful in creatingintelligent software systems that can deal withthe uncertainty and imprecision of real worldproblems

Some components of Intelligent systems are

• human-like - they possess human-likeexpertise within a specific domain,

• adaptable - they adapt themselves and learn todo better in a changing environment, and

• explanations - they explain how they makedecisions or take actions



8/28/200

2

Page 2

Soft Computing

3

How will we teach the techniques?How will we teach the techniques?

We will present

• multiple techniques from Soft Computing +,

• when each technique is applicable

• examples of industrial applications

“If the only tool you have is a hammer, thenevery problem looks like a nail”

- anonymous

Soft Computing

4

Soft ComputingSoft Computing

“Soft computing is an emergingapproach to computing whichparallels the remarkable ability ofthe human mind to reason andlearn in an environment ofuncertainty and imprecision”

- Lotfi Zadeh



8/28/200

Page 3

Soft Computing

5

Why Soft Computing?Why Soft Computing?Farmingcorn & cows

Manufacturingchairs & cars

Servicecontent and code

Industrial

Revolution

Information

Revolution

The information revolutiongoing on is allowing us toautomate informationprocessing tasks which requireintelligence much like theindustrial revolution automatedmanufacturing tasks

“Soft Computing” techniqueshave already been appliedsuccessfully.

Soft Computing

6

What is Soft Computing?What is Soft Computing?

Soft Computing is a field that currently includes

Fuzzy Logic

Neural Networks

Probabilistic Reasoning(Genetic Algorithms, BBN), and

Other related methodologies

• Case-Based Reasoning

Soft Computing combines knowledge,techniques, and methodologies from the sourcesabove to create intelligent systems



8/28/200

Page 4

Soft Computing

7

Fuzzy Logic - KaiFuzzy Logic - KaiSets with fuzzy boundaries

A = Set of tall people

Heights(cm)

170

1.0

Crisp set A

Membershipfunction

Heights(cm)

170 180

.5

.9

Fuzzy set A

1.0

Soft Computing

8

Fuzzy Set Theory - KaiFuzzy Set Theory - Kai

Fuzzy set theory provides a systematic calculusto deal with imprecise or incomplete information

Fuzzy if-then rules are used in fuzzy inferencesystems

If <1> is tall and <1> is athletic then <1> is goodbasketball player.

A B T-norm

X Y

w

A’ B’ C

Z



8/28/200

Page 5

Soft Computing

9

Neural Networks - KaiNeural Networks - KaiPattern matching technique where inputs arematched with a specific output pattern.

Network architecture

Modeled after the neurons in the brain.

Weights on the links

Learns by modifying the weights

x1

x2

y1

y2

Soft Computing

10

Genetic Algorithms - BillGenetic Algorithms - Bill

Use Idea of Evolution to Guide Search

HumanEvolution

Find Max ofa Function



8/28/200

Page 6

Soft Computing

11

Genetic Algorithms - BillGenetic Algorithms - BillAn optimization technique

10010110

01100010

101001001001100101111101

. . .

. . .

. . .

. . .

10010110

01100010

101001001001110101111001

. . .

. . .

. . .

. . .

Selection Crossover Mutation

Currentgeneration

Nextgeneration

Elitism

Soft Computing

12

Case-Based Reasoning - BillCase-Based Reasoning - Bill

A methodology ofsolving new problems

by adapting thesolutions of previoussimilar problems

Models the wayexperts reason using

their experience



8/28/200

Page 7

Soft Computing

13

What Is CBR? - Quiz

What is 12 x 12?

144

What is 12 x 13?

12 x 12 + 12

156

Soft Computing

14

Other Techniques – Bill & KaiOther Techniques – Bill & KaiBayesian belief networks

• represent and reason withprobabilistic knowledge

Decision Trees• classification using tree

structure

Least-squares estimator• statistical regression

Hybrid approaches• use multiple techniques

x<a

y<b y<c

ny

n y ny

z=f1 z=f2 z=f3 z=f4

Burglary

MaryCallsJohnCalls

Alarm

Earthquake

0.9 0.3

0.9 0.7

+3

+2

+1

0

0 1 2 3 4 5 6

y

x

y = x - 3



8/28/200

Page 8

Soft Computing

15

Soft Computing is a Hybrid MethodSoft Computing is a Hybrid Method

x1

x2

y1

y2

dog

dagKnowledge

baseAnimal?

dog

Neural

Character

Recognizer

Soft Computing

16

How does SC Relate to Other FieldsHow does SC Relate to Other Fields

What is AI?

“AI is the study of agents that exist in an environmentand perceive and act.” (S. Russell and P. Norvig)

“AI is the art of making computers do smart things.” (Waldrop)

“AI is a programming style, where programs operate on data

according to rules in order to accomplish goals.” (W. A. Taylor)

“AI is the activity of providing such machines as computerswith the ability to display behavior that would be regarded as

intelligent if it were observed in humans.” (R. McLoed)



8/28/200

Page 9

Soft Computing

17

How does SC Relate to Other FieldsHow does SC Relate to Other FieldsWhat is AI? (Jang)

The long term goal of AI research is thecreation and understanding of machine

intelligence

Conventional AI research focuses on anattempt to mimic human intelligent behavior

by expressing it in symbolic rules

BroadDefinition

Narrow

Definition

Soft Computing

18

How does SC Relate to Other FieldsHow does SC Relate to Other Fields

What is an Expert System (ES)?

User

KnowledgeEngineer

Knowledge

Acquisition

KBRules – if a then bFacts – a is true

Questions

Responses

InferenceEngine



8/28/200

1

Page 10

Soft Computing

19

Advantages Disadvantages

Functional Precise ReasoningProgramming Deterministic Learning

Symbolic Reasoning UncertaintyProgramming Learning Confidence(AI)

Soft Computing Uncertainty PreciseConfidence Deterministic

Types of ProgrammingTypes of Programming

Soft Computing

20

Stages of ReasoningStages of Reasoning

Complex Math

Logic

EvolutionExperienceUncertainty

H um a n s

C om p u t e r s

FunctionalProgramming

SymbolicProgramming(AI)

Soft Computing



8/28/200

1

Page 11

Soft Computing

21

Soft Computing CharacteristicsSoft Computing CharacteristicsHuman Expertise (if-then rules, cases,

conventional knowledge representations)

Biologically inspired computing models (NN)

New optimization techniques (GA, simulated annealing)

Model-free learning (NN, CBR)

Fault tolerance (deletion of neuron, rule, or case)

Real-world applications (large scale with uncertainties)

Soft Computing

22

Soft Computing EntertainmentSoft Computing EntertainmentStar Trek

Kirk and Spock are the classic fuzzyand crisp reasoners

Errand of Mercy episode• Klingon army attacks a neutral planet• Kirk and Spock beam down

• Enterprise is chased away• Inhabitants are not concerned• Should they try and help?

Kirk SpockFuzzy Crisp

Spock – odds of succeeding are 7,249.5 to 1

Kirk – we should do it anyway



8/28/200

12

Page 12

Soft Computing

23

Soft Computing in HistorySoft Computing in History“If a man will begin with certainties,

he will end with doubts,but if he will be content to begin with doubts,he shall end in certainties.”

- Francis Bacon 1605THE ADVANCEMENT OF LEARNING

Documents

Chapter 1 Sc