Interval Type-2 Fuzzy logic vs Perceptual Computing

Post on 01-Jun-2018

215 views 0 download

Transcript of Interval Type-2 Fuzzy logic vs Perceptual Computing

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 1/27

Interval Type-2 Fuzzy Logic

System versus Perceptual

Computer: Similarities and

Differences

Jerry M. Mendel

University of Southern California

Los Angeles, CA

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 2/27

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 3/27

IT2 FLS vs Per-C: Issues

•Inputs

•Fuzzifier 

•Rules

Inference•Output Processing

•Outputs

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 4/27

IT2 FLS vs Per-C: Issues

•Inputs

•Fuzzifier 

•Rules

Inference•Output Processing

•Outputs

•Inputs

•Encoder 

•CWW Engine

Output of CWW Engine•Decoder 

•Recommendation + Data

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 5/27

IT2 FLS vs Per-C: Applications

•“Function Approximation”

•Fuzzy logic control

•Signal processing

Rule-based classification

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 6/27

IT2 FLS vs Per-C: Applications

•“Computing With Words”

•Investment advising

•Social judgments

Decision making

•“Function Approximation”

•Fuzzy logic control

•Signal processing

Rule-based classification

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 7/27

IT2 FLS vs Per-C

•Inputs

•Numbers first, then

the Membership

Functions (MFs)

•Doesn’t matter whatyou call the fuzzy

sets

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 8/27

•Inputs

•Words first, then

the MFs

•Words that

mean somethingto end-user label

the fuzzy sets

IT2 FLS vs Per-C

•Inputs

•Numbers first, then

the Membership

Functions (MFs)

•Doesn’t matter whatyou call the fuzzy

sets

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 9/27

•Fuzzifier—Different

kinds (choices)•Singleton

•T1 FS—a fuzzy

number • IT2 FS—a fuzzy-fuzzy

number 

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 10/27

•Encoder 

•Words mean different things

to different people

• IT2 FS—No choice

Data from group of subjects• IA maps data into an FOU

• Three canonical FOUs

• Codebook {Wi, FOU(Wi)}

IT2 FLS vs Per-C

•Fuzzifier—Different

kinds (choices)•Singleton

•T1 FS—a fuzzy

number 

• IT2 FS—a fuzzy-fuzzy

number 

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 11/27

•Rules—IF-THEN

•From experts

•From data

• Independent of kind of 

FSs used

•Words in antecedents

and consequents

modeled as IT2 FSs

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 12/27

•Rules—IF-THEN

•From experts

•From data

• Independent of kind of 

FSs used

•Words in antecedents

and consequents

modeled as IT2 FSs

•CWW Engine

• IF-THEN rules

•LWA

•Others under development

• All words used by the CWW

Engine must be in a

Codebook

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 13/27

•Inference

•Mamdani

• Extended sup-star composition:

firing interval

• Computations only involve

LMFs and UMFs

• Fired rule outputs may be

combined or not, depending on

kind of output processing

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 14/27

•Inference

•TSK

• Firing interval

• Computations only involve

LMFs and UMFs

• Fired rule outputs combined

using TSK formula

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 15/27

•Inference

•Mamdani

•TSK

IT2 FLS vs Per-C

•Output of CWW Engine

• IF-THEN rules

• Similarity used to compute firing

level

• Perceptual Reasoning used to

aggregate fired rules

• Resulting output IT2 FS

resembles word FOUs—new

requirement for CWW

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 16/27

•Perceptual Reasoning

IT2 FLS vs Per-C

•Output of CWW Engine

• IF-THEN rules

• Similarity used to compute firing

level

• Perceptual Reasoning used to

aggregate fired rules

• Resulting output IT2 FS

resembles word FOUs—new

requirement for CWW

Y PR   =

 f iGii=1

 M 

 f  j  j =1

 M 

=

 f i

 f  j  j =1

 M 

Gii=1

 M 

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 17/27

•Output of CWW Engine

•LWA (words, T1 FSs,intervals and numbers)

• Extension Principle

• Alpha-cuts function

decomposition theorem

• Two FWAs

• IWAs

• FOU(LWA) resembles word

FOUs

IT2 FLS vs Per-C

•Inference

•Mamdani

•TSK

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 18/27

•Output of CWW Engine

•LWA (words, T1 FSs,intervals and numbers)

• Extension Principle

• Alpha-cuts function

decomposition theorem

• Two FWAs

• IWAs

• FOU(LWA) resembles word

FOUs

IT2 FLS vs Per-C

•LWA

 

Y  LWA   =

 X iW ii=1

 M 

W  j  j =1

 M 

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 19/27

•Output Processing

•Type-reduction

• Different kinds

• KM algorithms

• TR set is an IVFS—uncertainty

measure

•Defuzzification

• Average of TR FS

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 20/27

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 21/27

•Outputs

• Crisp outputs that are used in anaction

• TR IVFS that can be used as a

measure of uncertainties that have

flowed through the IT2FLS

(analogous to a confidence interval)

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 22/27

•Outputs

• Crisp outputs that are used in anaction

• TR IVFS that can be used as a

measure of uncertainties that have

flowed through the IT2FLS

(analogous to a confidence interval)

•Recommendation + Data

• People want to know “Why?”

• Linguistic and numerical outputs

• Centroid and ranking bands can be

used as measures of uncertainties

that have flowed through the Per-C

IT2 FLS vs Per-C

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 23/27

•Words before FSs

Words mean different things to different people•Words mean similar things to different people

•Words or a mixture of words and numbers always excite the Per-C

•CWW Engines are constrained so that their outputs resemble the

FOUs in the Codebook

•Computations developed for IT2 FLSs are used in Perceptual

Computing

•Similarity, rank and subsethood are important in Per-C

IT2 FLS vs Per-C:

Recapitulation

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 24/27

Conclusions

•There are many differences between anIT2 FLS and a Perceptual Computer 

• By comparing their architectures, block-

by-block, it is easy to enumerate thosedifferences

They are used for very different kinds of problems

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 25/27

IT2 FLS vs Per-C

•One Reference

J. M. Mendel, Uncertain Rule-Based Fuzzy Logic Systems:

Introduction and New Directions,

Prentice-Hall, 2001

•There are now a multitude of references for IT2 FLSs

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 26/27

IT2 FLS vs Per-C

•Reference

J. M. Mendel, Uncertain Rule-Based Fuzzy Logic Systems:

Introduction and New Directions,

Prentice-Hall, 2001

•There are now a multitude ofreferences for IT2 FLSs

•Reference

J. M. Mendel and D. Wu,Perceptual Computing: Aiding 

People in Making Subjective

Judgments, Wiley and IEEE Press,

2010•There is not yet a multitude of 

references for Perceptual

Computing

8/9/2019 Interval Type-2 Fuzzy logic vs Perceptual Computing

http://slidepdf.com/reader/full/interval-type-2-fuzzy-logic-vs-perceptual-computing 27/27

Thanks