Fuzzy Rule-Based Models
description
Transcript of Fuzzy Rule-Based Models
-
11 Fuzzy Rule-BasedModels
Fuzzy Systems EngineeringToward Human-Centric Computing
-
11.1 Fuzzy rules as a vehicle of knowledge representation
11.2 General categories of fuzzy rules and their semantics
11.3 Syntax of fuzzy rules
11.4 Basic functional modules
11.5 Types of rule-based systems and architectures
11.6 Approximation properties of fuzzy rule-based models
11.7 Development of rule-based systems
Contents
Pedrycz and Gomide, FSE 2007
-
11.8 Parameter estimation procedure for functionalrule-based systems
11.9 Design of rule-based systems: consistency,completeness and the curse of dimensionality
11.10 Course of dimensionality in rule-based systems
11.11 Development scheme of fuzzy rule-based models
Pedrycz and Gomide, FSE 2007
-
11.1 Fuzzy rules as a vehicle ofknowledge representation
Pedrycz and Gomide, FSE 2007
-
Rule conditional statement
If input variable is A then output variable is B
A and B: descriptors of pieces of knowledge
rule: expresses a relationship between inputs and outputs
Example
If the temperature is high then the electricity demand is high
If and then parts ....... formed by information granules
sets rough sets fuzzy sets
Pedrycz and Gomide, FSE 2007
-
Rule-based system/model (FRBS)
FRBS is a family of rules of the form
If input variable is Ai then output variable is Bi
i = 1, 2,..., c
Ai and Bi are information granules
More complex rules
If input variable1 is Ai and input variable2 is Bi and ..... then output variable is Zi
multidimensional input space (Cartesian product of inputs) individual inputs aggregated by the and connective highly parallel, modular granular model
Pedrycz and Gomide, FSE 2007
-
11.2 General categories of fuzzyrules and their semantics
Pedrycz and Gomide, FSE 2007
-
Multi-input multi-output fuzzy rules
If X1 is A1 and X2 is A2 and ..... and Xn is Anthen Y1 is B1 and Y2 is B2 and ..... and Ym is Bm
Xi = variables whose values are fuzzy sets AiYj = variables whose values are fuzzy sets Bj
Ai on Xi, i = 1, 2,...,n
Bj on Yj, j = 1, 2,...,m
No loss of generality if we assume rules of the form
If X is A and Y is B then Z is C
Pedrycz and Gomide, FSE 2007
-
Certainty-qualified rules
If X is A and Y is B then Z is C with certainty
[0,1]
: degree of certainty of the rule
= 1 rule is certain
Pedrycz and Gomide, FSE 2007
-
Gradual rules
the more X is A the more Y is B
relationships between changes in X and Y
captures tendency between information granules
Examples:
the higher the income, the higher the taxes
the lower the temperature, the higher energy consumption
Pedrycz and Gomide, FSE 2007
-
Functional fuzzy rules
If X is Ai then y = f (x,ai)
f : X Y
xRn
Rule: confines the function to the support of granule Ai
f : linear or nonlinear (neural nets, etc..)
Highly modular models
Pedrycz and Gomide, FSE 2007
-
11.3 Syntax of fuzzy rules
Pedrycz and Gomide, FSE 2007
-
Backus-Naur form (BNF)
If_then_rule ::= if antecedent then consequent{certainty}gradual_rule ::= word antecedentword consequent
word ::= more {less}antecedent ::= expressionconsequent ::= expressionexpression ::= disjunction{and disjunction}disjunction ::= variable{orvariable}
variable ::= attribute is valuecertainty ::= none{certainty [0,1]}
Pedrycz and Gomide, FSE 2007
-
Construction of computable representations
Main steps:
1. specification of the fuzzy variables to be used
2. association of the fuzzy variables using fuzzy sets
3. computational formalization of each rule using fuzzyrelations and definition of aggregation operator to combinerules together
Pedrycz and Gomide, FSE 2007
-
11.4 Basic functionalmodules of FRBS
Pedrycz and Gomide, FSE 2007
-
I
n
p
u
t
I
n
t
e
r
f
a
c
e
Rule Base
Data Base
Fuzzy Inference O
u
t
p
u
t
I
n
t
e
r
f
a
c
e
X Y
General architecture of FRBS
Fuzzy if-then rules(input-output relationship)
Parametersof the FRBS
Process inputs and rules(approximate reasoning)
Pedrycz and Gomide, FSE 2007
-
Input interface
(attribute) of (input) is (value)
the temperature of the motor is high
Canonical (atomic) form
p: X is A temperature (motor) is highX A
fuzzy set
Low Medium High
x (C)Pedrycz and Gomide, FSE 2007
-
Multiple fuzzy inputs: conjunctive canonical form
p : X1 is A1 and X2 is A2 and ..... and Xn is An conjunctive canonical form
Xi are fuzzy (linguistic) variables
Ai : fuzzy sets on Xi
i = 1, 2, ..., n
Compound proposition induces a fuzzy relation P on X1X1... Xn
)()()()()(1
221121 iin
innn xATxtAtxtAxAx,,x,xP
=
== KK
p : (X1, X2 , ....., Xn) is P
t (T) = t-norm
Pedrycz and Gomide, FSE 2007
-
0 5 100
10
20
x
y
(b) Contours of P
Example Fuzzy relation associated with (X,Y) is P
Triangular fuzzy sets A1(x,4,5,6) = A, A2(y,8,10,12) = B
t-norm: algebraic product
P
Pedrycz and Gomide, FSE 2007
-
q : X1 is A1 or X2 is A2 or ..... or Xn is An disjunctive canonical form
Xi are fuzzy (linguistic) variables
Ai : fuzzy sets on Xi
i = 1, 2, ..., n
Compound proposition induces a fuzzy relation Q on X1X1... Xn
)()()()()(1
221121 iin
innn xASxsAsxsAxAx,,x,xQ
=
== KK
q : (X1, X2 , ....., Xn) is Q
s (S) = t-conorm
Multiple fuzzy inputs: disjunctive canonical form
Pedrycz and Gomide, FSE 2007
-
Example Fuzzy relation associated with (X,Y) is Q
Triangular fuzzy sets A1(x,4,5,6) = A, A2(y,8,10,12) = B
t-conorm: probabilistic sum
Q
0 5 100
10
20
x
y
(d) Contours of Q
Pedrycz and Gomide, FSE 2007
-
Rule base
Fuzzy rule: If X is A then Y is B relationship between X and Y
Semantics of the rule is given by a fuzzy relation R on XY
R determined by a relational assignment
R(x,y) = f (A(x),B(y)) (x, y)XY
f : [0,1]2 [0,1]
In general f can be
fuzzy conjunction: ft fuzzy disjunction: fs fuzzy implication: fi
Pedrycz and Gomide, FSE 2007
-
Choose a t-norm t and define:
R(x,y) ft (x,y) = A(x) t B(y) (x,y) XY
Examples:
t = min
Rc(x,y) fc (x,y) = min[A(x) t B(y)] (Mamdani)
t = algebraic product
Rp(x,y) fp (x,y) = A(x)B(y) (Larsen)
Fuzzy conjunction
Pedrycz and Gomide, FSE 2007
-
Rc(x,y) = min {A(x), B(y)}
(A(x), B(y))[0,1]2Rc(x,y) = min {A(x), B(y)}
A(x) = A(x,4,5,6)B(y) = B(y,4,5,6)
Example: t = min
Pedrycz and Gomide, FSE 2007
-
Rp(x,y) =A(x)B(y)
(A(x), B(y))[0,1]2Rp(x,y) = A(x)B(y)A(x) = A(x,4,5,6)B(y) = B(y,4,5,6)
Example: t = algebraic product
Pedrycz and Gomide, FSE 2007
-
Choose a t-conorm s and define:
Rs(x,y) fs (x,y) = A(x) s B(y) (x,y) XY
Examples:
s = max
Rm(x,y) fm (x,y) = max[A(x) t B(y)]
s = Lukasiewicz t-conorm
Rl (x,y) fl (x,y) = min[1, A(x) + B(y)]
Fuzzy disjunction
Pedrycz and Gomide, FSE 2007
-
Example: s = max
Rl (x,y) = max{A(x), B(y)}
(A(x), B(y))[0,1]2Rl (x,y) = max {A(x), B(y)}
A(x) = A(x,4,5,6)B(y) = B(y,4,5,6)
Pedrycz and Gomide, FSE 2007
-
Example: s = Lukasiewicz
Rc(x,y) = min{1, A(x)+B(y)}
(A(x), B(y))[0,1]2Rc(x,y) = min{1, A(x)+B(y)}
A(x) = A(x,4,5,6)B(y) = B(y,4,5,6)
Pedrycz and Gomide, FSE 2007
-
Fuzzy implication
Choose a fuzzy implication fi and define:
Ri(x,y) fi (x,y) (x,y) XY
fi : [0,1]2 [0,1] is a fuzzy implication if:
1. B(y1) B(y2) fi (A(x), B(y1)) fi (A(x), B(y2)) monotonicity 2nd argument
2. fi (0, B(y)) = 1 dominance of falsity
3. fi (1, B(y)) = B(y) neutrality of truth
Pedrycz and Gomide, FSE 2007
-
Further requirements may include:
4. A(x1) A(x2) fi (A(x1), B(y)) fi (A(x2), B(y)) monotonicity 1st argument
5. fi (A(x1), fi (A(x2), B(y)) = fi (A(x2), fi (A(x1), B(y)) exchange
6. fi (A(x), A(x)) = 1 identity
7. fi (A(x), B(y)) = 1 A(x) B(y) boundary condition
8. fi is a continuous function continuity
Pedrycz and Gomide, FSE 2007
-
+++
= )(1)()1()(11min))()((
xAyBxA
,yB,xAf
]))()(1(1[min))()(( 1 w/www yBxA,yB,xAf +=
=
otherwise0)()(if1))()(( yBxAyB,xAfa
=
otherwise)()()(if1))()((
yByBxA
yB,xAfg
=otherwise)(
)()()(if1
))()((xAyB
yBxAyB,xAfe
)]()(1[max))()(( yB,xAyB,xAfb =
Name Definition Comment
Lukasiewicz fl(A(x), B(y)) = min [1, 1 A(x) + B(y)]
Pseudo-Lukasiewicz > -1
Pseudo-Lukasiewicz w > 0
Gaines
Gdel
Goguen
Kleene
Reichenbach
Zadeh
Klir-Yuan
)]()()(1))()(( yBxAxAyB,xAfr +=
))]()((min)(1[max))()(( yB,xA,xAyB,xAfz =
)()()(1))()(( 2 yBxAxAyB,xAfk +=
Examples of fuzzy implications
Pedrycz and Gomide, FSE 2007
-
Example: fl = Lukasiewicz
Rl (x,y) = min{1, 1 A(x)+B(y)}
(A(x), B(y))[0,1]2Rl (x,y) = min{1, 1 A(x)+B(y)}
A(x) = A(x,4,5,6)B(y) = B(y,4,5,6)
Pedrycz and Gomide, FSE 2007
-
Example: fk = KlirYuan
Rk (x,y) = 1 A(x)+A(x)2B(y)
(A(x), B(y))[0,1]2Rk (x,y) = 1 A(x)+A(x)2B(y)
A(x) = A(x,4,5,6)B(y) = B(y,4,5,6)
Pedrycz and Gomide, FSE 2007
-
Categories of fuzzy implications:
1. s-implications
zLukasiewic)}()(11min{))(),((
Kleene)]()(1max[))(),((
)()()())(),((
yBxA,yBxAf
yB,xAyBxAf
y,xysBxAyBxAf
g
b
is
+=
=
= YX
2. r-implications
Gdel)()()()()(if1))(),((
min
)()]()(|]10[sup[))(),((
>
=
=
=
yBxAyByBxA
yBxAf
t
y,xyBctxA,cyBxAf
g
ir YX
Pedrycz and Gomide, FSE 2007
-
Semantics of gradual rules
x
A(x)
B(y)
y
1.0 1.0
BRd
the more X is A, the more Y is B B(y) A(x) xX and yY
BRd = {yY |B(y) A(x)} for each xX
Pedrycz and Gomide, FSE 2007
-
Example: Rd = fa = Gaines
=
otherwise0)()(if1),( xAyByxRd
Rd(x,y) (A(x), B(y))[0,1]2
Rd (x,y)A(x) = A(x,3,5,7)B(y) = B(y,3,5,7)
00.2
0.40.6
0.81
0
0.5
10
0.2
0.4
0.6
0.8
1
A(x)
(a) Gradual rule Rd = fa
B(y)
Pedrycz and Gomide, FSE 2007
-
Main types of rule bases
Fuzzy rule base {R1, R2,....,RN} finite family of fuzzy rules
Fuzzy rule base can assume various formats:
1. fuzzy graph
Ri: If X is Ai then Y is Bi is a fuzzy granule in XY, i = 1,...,N
2. fuzzy implication rule base
Ri: If X is Ai then Y is Bi is fuzzy implication, i = 1,...,N
3. functional fuzzy rule base
Ri: If X is Ai then y = fi(x) is a functional fuzzy rule, i = 1,...,N
Pedrycz and Gomide, FSE 2007
-
Fuzzy graph
Fuzzy rule base R collection of rules R1, R2,....,RN
Each fuzzy rule Ri is a fuzzy granule (point)
Fuzzy graph R is a collection of fuzzy granules
granular approximation of a function
R = R1 or R2 or....or RN
general form
)(11
iN
ii
N
ii BARR ==
==
UU
)]()([),(1
ytBxASyxR iiN
i==
Pedrycz and Gomide, FSE 2007
-
Point
0 2 4 6 8 100
2
4
6
8
10
x
(b)
P
0 2 4 6 8 100
1
x
A
(
x
)
(c)
A
0
2
4
6
8
10
01
(a)
B(y)
By
Point P in XYP = AB
A is a singleton in XB is a singleton in Y
Pedrycz and Gomide, FSE 2007
-
Granule
0 2 4 6 8 100
2
4
6
8
10
xy
(b)
G
0 2 4 6 8 100
1
x
A
(
x
)
(c)
A
0
2
4
6
8
10
01
(a)
B(y)
B
Granule G in XYG = AB
A is an interval in XB is an interval in Y
Pedrycz and Gomide, FSE 2007
-
Fuzzy granules fuzzy points
0 5 100
2
4
6
8
10
x
y
R
(b) Fuzzy granule R
0 2 4 6 8 100
1
x
A
(
x
)
(c)
A
0
2
4
6
8
10
01
(a)
B(y)
B
fuzzy granule R in XYR = AB
A is a fuzzy set on XB is a fuzzy set on Y
Pedrycz and Gomide, FSE 2007
-
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
(b) Fuzzy granules Ri
R1
R2
R3
R4
R5
y
0 2 4 6 8 100
1
x
(c)
A1 A2 A3 A4 A5
0
2
4
6
8
10
01
(a)
B(y)
B1
B2
B3
B4
B5
y
Fuzzy rule base as a set fuzzy granules
Ri = AiBi
Pedrycz and Gomide, FSE 2007
-
0 2 4 6 8 10 120
2
4
6
8
10
12
x
y
(a) function y = f(x)
f
1 2 3 4 5 6 7 8 9 10 11 12
1
2
3
4
5
6
7
8
9
10
11
12
x
y
(b)Granular approximation of y = f(x)
R
R1
R2
R3R4
R5 R6 R7R8
Graph of a function f and its granular approximation R
f R
Ri = AiBi
Pedrycz and Gomide, FSE 2007
-
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
(b)Fuzzy rule base as a fuzzy graph (t = min)
R = Union Ri
y
Fuzzy rule base and fuzzy graphExample 1
Ri = AiBi Ri(x,y) = min [Ai(x), Bi(y)]
R = Ri R(x,y) = max [Ri(x,y), i = 1,..., N ]Pedrycz and Gomide, FSE 2007
-
Fuzzy rule base and fuzzy graphExample 2
Ri = Ai t Bi Ri(x,y) = Ai(x) Bi(y)
R = Ri R(x,y) = max [Ri(x,y), i = 1,..., N ]
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
(d) Fuzzy rule base as a fuzzy graph (t = product)
R = Union Ri
y
Pedrycz and Gomide, FSE 2007
-
Fuzzy implication
Fuzzy rule base R collection of rules R1, R2,....,RN
Each fuzzy rule Ri is a fuzzy implication
Fuzzy rule base R is a collection of fuzzy relations
relation R is obtained using intersection
R = R1 and R2 and....and RN
general form
)(111
iN
ii
N
ii
N
ii BAfRR III
===
===
))()((1
yB,xAfTR iiiN
i==
Pedrycz and Gomide, FSE 2007
-
Fuzzy rule as an implication
0 2 4 6 8 100
2
4
6
8
10
x
y
(b) Lukasiewicz implication R
R
0 2 4 6 8 100
1
x
A
(
x
)
(c)
A
0
2
4
6
8
10
01
(a)
B(y)
B
fuzzy rule R in XYR = fl (A,B)Lukasiewicz implication
Pedrycz and Gomide, FSE 2007
-
Fuzzy rule base and fuzzy implicationExample 1a
Ri = fl (A,B) Ri(x,y) = min [1, 1 Ai(x) + Bi(y)] Lukasiewicz implication
R = Ri R(x,y) = min [Ri(x,y), i = 1,..., 5] min t-norm
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
y
(b) Fuzzy rule base as Lukasiewicz implication (t = min)
Pedrycz and Gomide, FSE 2007
-
Fuzzy rule base and fuzzy implicationExample 1b
Ri = fl (A,B) Ri(x,y) = min [1, 1 Ai(x) + Bi(y)] Lukasiewicz implication
R = Ri R(x,y) = R1(x,y) tl R2(x,y) tl .... tl Ri(x,y) Lukasiewicz t-norm
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
y
(b) Fuzzy rule base as Lukasiewicz implication (t = min)
Pedrycz and Gomide, FSE 2007
-
Fuzzy rule base and fuzzy implicationExample 2a
Ri = fz (A,B) Ri(x,y) = max [1 Ai(x), min(Ai(x), Bi(y)] Zadeh implicationR = Ri R(x,y) = min [Ri(x,y), i = 1,..., 5] min t-norm
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
y
(b) Fuzzy rule base as Zadeh implication (t = min)
Pedrycz and Gomide, FSE 2007
-
Fuzzy rule base and fuzzy implicationExample 2b
Ri = fz (A,B) Ri(x,y) = max [1 Ai(x), min(Ai(x), Bi(y)] Zadeh implicationR = Ri R(x,y) = R1(x,y) tl R2(x,y) tl .... tl Ri(x,y) Lukasiewicz t-norm
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
x
y
(d) Fuzzy rule base as Zadeh implication (t = Lukasiewicz)
Pedrycz and Gomide, FSE 2007
-
Data base
Data base contains definitions of: universes scaling functions of input and output variables granulation of the universes membership functions
Granulation granular constructs in the form of fuzzy points granules along different regions of the universes
Construction of membership functions expert knowledge learning from data
Pedrycz and Gomide, FSE 2007
-
Granulation
X X
granular constructs inthe form of fuzzy points
granules along differentregions of the universes
Pedrycz and Gomide, FSE 2007
-
Fuzzy inference
Basic idea of inference
0 2 4 6 8 10 120
2
4
6
8
10
12
x
y
(a)
I
ac
f
a
b
x = a
y = f (x) y = b
b = ProjY (ac f )
b = ProjY ( I )
Pedrycz and Gomide, FSE 2007
-
Inference involves operations with sets
x = A y = f (x) B = f (A) ={f (x), xA}
B = ProjY (Ac f )
B = ProjY ( I ) 0 2 4 6 8 10 1202
4
6
8
10
12
x
y
(b)
I
Ac
A
a b
B c
d
f
Pedrycz and Gomide, FSE 2007
-
Inference involving sets and relations
x is A(x,y) is Ry is B
B = ProjY (Ac R )
B = ProjY ( I )
f
0 2 4 6 8 10 120
2
4
6
8
10
12
x
y
(a)
I
Ac
R
A
B
Pedrycz and Gomide, FSE 2007
-
Fuzzy inference ands operations with fuzzy sets and relations
X is A (fuzzy set on X)(X,Y) is R (fuzzy relation on XY)Y is B (fuzzy set on Y)
B = ProjY (Ac R )
B = ProjY ( I )
f
2 4 6 8 10 12
2
4
6
8
10
12
x
y
(b)
A
B R
I
Ac
)}()({sup)( y,xtRxAyBx X
=
Pedrycz and Gomide, FSE 2007
-
Fuzzy inference
Compositional rule of inference
X is A (X,Y) is RY is B
X is A(X,Y) is RY is AoR
RAB o=
Pedrycz and Gomide, FSE 2007
-
procedure FUZZY-INFERENCE (A, R) returns a fuzzy setinput : fuzzy relation: R
fuzzy set: Alocal: x, y: elements of X and Y
t: t-norm
for all x and y doAc(x,y) A(x)
for all x and y doI(x,y) Ac(x,y) t R(x,y)
B(y) supx
I(x,y) return B
Fuzzy inference procedure
Pedrycz and Gomide, FSE 2007
-
Example: compositional rule of inference
Pedrycz and Gomide, FSE 2007
-
Example: fuzzy inference with fuzzy graph
Pedrycz and Gomide, FSE 2007
-
11.5 Types of rule-basedsystems and architectures
Pedrycz and Gomide, FSE 2007
-
Linguistic fuzzy models
P: X is A and Y is B input
R1: If X is A1 and Y is B1 then Z is C1......................
Ri: If X is Ai and Y is Bi then Z is Ci rule base.......................
RN: If X is AN and Y is BN then Z is CN
Z: Z is C output
all fuzzy sets A, B, Ai,s and Bi,s are given rule and connectives (and, or) with known semantics membership function of fuzzy set C = ??
Pedrycz and Gomide, FSE 2007
-
min-max fuzzy models
Assume
P: X is A and Y is B P(x,y) = min{A(x), B(y)}
Ri: If X is Ai and Y is Bi then Z is Ci Ri(x,y,z) = min{Ai(x), Bi(y), Ci(z)}
i = 1,..., N
)]}1),(max(),({min[sup)(
1
,...,Niz,y,xRy,xPzC
RPRPC
iy,x
N
ii
==
==
=
Uoo
Using the compositional rule of inference (t = min)
Pedrycz and Gomide, FSE 2007
-
rulethiofactivationofdegreetheis
}1),(max{}1)max{()(
)()(
)Poss()]()([sup
)Poss()]()([sup
)]}()()()()({sup)]}(),({min[sup)(
)(111
====
=
==
==
==
=
====
===
i
iiiii
iiii
iiiy
iiix
iiiy,x
iy,x
i
ii
N
ii
N
ii
N
ii
N,,izCN,,i,CnmzC
zCnmzC
nB,ByByB
mA,AxAxA
zCyBxAyBxAz,y,xRy,xPzC
RPC
CRPRPRPC
KK
o
ooo UUU
Pedrycz and Gomide, FSE 2007
-
procedure MIN-MAX-MODEL (A,B) returns a fuzzy setlocal: fuzzy sets: Ai, Bi, Ci, i =1,.., N
activation degrees: i
Initialization C =
for i = 1: N domi = max (min (A, Ai))ni = max (min (B, Bi))i = min (mi, ni)if i 0 then Ci = min (i , Ci) and C = max(C, Ci)
return C
min-max fuzzy model processing
Pedrycz and Gomide, FSE 2007
-
Example: min-max fuzzy model processing
Ai Aj Bi Bj A B
mi
mj
ni
nj
Ci Cj 1 1 1
nj
mi Ci
Cj
x y z
Pedrycz and Gomide, FSE 2007
-
min-max fuzzy model architecture
Poss 1 Min
Max
A1,B1 C1
Poss i Min
Ai,Bi Ci
Poss N Min
AN,BN CN
C (A,B)
Ci
CN
C1
Pedrycz and Gomide, FSE 2007
-
Special case: numeric inputs
=
=
=
=
otherwise0if1)(
otherwise0if1)( oo yyyBandxxxA
Numeric output
iN
iii
N
iiii
v
nm
vnm
z
dzzCdzzzC
z
valuesmodalaverageweighted)(
)(
ationdefuzzificcentroid)()(
1
1
=
=
=
=
Z
Z
Pedrycz and Gomide, FSE 2007
-
ExampleP: X is xo and Y is yo inputs (xo, yo), xo, yo [-2, 2]
R1: If X is A1 and Y is B1 then Z is C1rules
R2: If X is A2 and Y is B2 then Z is C2
N = 2, centroid defuzzification
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
0.5
1
x
A
i
(
x
)
A1 A2
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
0.5
1
y
B
i
(
y
)
B2 B2
(a) Input and output fuzzy sets
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
0.5
1
z
C
i
(
z
)
C1 C2
-2-1
01
2
-2
-1
0
1
2
-0.5
0
0.5
x
(b) Input-output mapping
y
z
Pedrycz and Gomide, FSE 2007
-
min-sum fuzzy models
Assume
P: X is A and Y is B P(x,y) = min{A(x), B(y)}
Ri: If X is Ai and Y is Bi then Z is Ci Ri(x,y,z) = min{Ai(x), Bi(y), Ci(z)}
i = 1,..., N
=
=
=
N
iii
iiiy,x
i
CwzC
zCyBxAyBxAzC
1)(
)]()()()()([sup)(
Using the compositional rule of inference (t = min)
Additive fuzzy models(Kosko, 1992)
Pedrycz and Gomide, FSE 2007
-
Example: min-sum fuzzy model processing
Ai Aj Bi Bj A B
mi
mj
ni
nj
Ci Cj 1 1 1
nj
mi Ci Cj
x y z
Ci
Pedrycz and Gomide, FSE 2007
-
min-sum fuzzy model architecture
Poss 1 Min
A1,B1 C1
Poss i Min
Ai,Bi Ci
Poss N Min
AN,BN CN
C (A,B) wi
w1
wN
CN
C1
Ci
Pedrycz and Gomide, FSE 2007
-
ExampleP: X is xo and Y is yo inputs (xo, yo), xo, yo [-2, 2]
R1: If X is A1 and Y is B1 then Z is C1rules
R2: If X is A2 and Y is B2 then Z is C2
N = 2 w1 = w2 = 1, centroid defuzzification
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
0.5
1
x
A
i
(
x
)
A1 A2
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
0.5
1
y
B
i
(
y
)
B2 B2
(a) Input and output fuzzy sets
-2 -1.5 -1 -0.5 0 0.5 1 1.5 20
0.5
1
z
C
i
(
z
)
C1 C2
-2-1
01
2
-2
-1
0
1
2
-0.5
0
0.5
x
(b) Input-output mapping
y
z
Pedrycz and Gomide, FSE 2007
-
product-sum fuzzy models
1- Productprobabilistic sum
)()(
)()(
1zCSzC
zCnmzC
iN
ip
iiii
=
=
=
2- Productsum
=
=
=
N
ii
iiii
zCzC
zCnmzC
1)()(
)()(
-2-1
01
2
-2
-1
0
1
2
-0.5
0
0.5
x
(b) Input-output mapping of product-probabilistic sum model
y
z
-2-1
01
2
-2
-1
0
1
2
-0.5
0
0.5
x
(d) Input-output mapping of product-sum model
y
z
Pedrycz and Gomide, FSE 2007
-
3 - Bounded product-bounded sum
]10[
}1min{
}10max{
)()(
)()(
1
,a,b
ba,ba
ba,ba
zCzC
zCnmzC
iN
i
iiii
+=
+=
=
=
=
-2-1
01
2
-2
-1
0
1
2
-1
-0.5
0
0.5
1
x
(c) Input-output mapping of bounded product-bounded sum model
y
z
Pedrycz and Gomide, FSE 2007
-
Functional fuzzy models
P: X is x and Y is y input
R1: If X is A1 and Y is B1 then z = f1 (x,y)......................
Ri: If X is Ai and Y is Bi then z = fi (x,y) rule base.......................
RN: If X is AN and Y is BN then z = fN (x,y)
i(x,y) = Ai(x) t Bi(y) t = t-norm degree of activation
output
=
=
== N
ii
ii
N
iii
y,x
w,y,xfy,xwz
1
1 )()()(
Pedrycz and Gomide, FSE 2007
-
Functional fuzzy model architecture
w1
f1(x,y)
A1,B1
wi
Ai,Bi
wN
AN,BN
z (x,y)
f2(x,y)
fN(x,y)
Pedrycz and Gomide, FSE 2007
-
Example 1P: X is x inputs x [0, 3]
R1: If X is A1 then z = xrules
R2: If X is A2 then z = x + 3
0 0.5 1 1.5 2 2.5 30
0.2
0.4
0.6
0.8
1
1.2
x
A
i A1 A2
(a) Antecedent fuzzy sets
0 0.5 1 1.5 2 2.5 30
0.5
1
1.5
2
2.5
3
f1f2
x
y
(b) Consequent functions
Pedrycz and Gomide, FSE 2007
-
+
++
=
)32[if3]21[if)3)(()(]10(if
21,xx
,xxxAxxA,xx
z
0 0.5 1 1.5 2 2.5 30
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
x
y
(b) Output of the functional model
output
Pedrycz and Gomide, FSE 2007
-
Example 2P: X is x inputs x [0, 3]
R1: If X is A1 then y = sin(2x)
R2: If X is A2 then y = 0.5x rules
R3: If X is A3 then y = sin(3x)
-5 -4 -3 -2 -1 0 1 2 3 4 50
0.2
0.4
0.6
0.8
1
1.2
A1 A2 A3
(a) Antecedent fuzzy sets
x
y
-5 -4 -3 -2 -1 0 1 2 3 4 5-1.5
-1
-0.5
0
0.5
1
1.5(b) output of the functional fuzzy model
y
x
output
Pedrycz and Gomide, FSE 2007
-
Example 2P: X is x inputs x [0, 3]
R1: If X is A1 then y = 1
R2: If X is A2 then y = x rules
R3: If X is A3 then y = 1
-5 -4 -3 -2 -1 0 1 2 3 4 50
0.2
0.4
0.6
0.8
1
1.2
A1 A2 A3
(a) Antecedent fuzzy sets
x
y
-5 -4 -3 -2 -1 0 1 2 3 4 5-1.5
-1
-0.5
0
0.5
1
1.5
x
y
(c) output of the functional fuzzy model
output
Pedrycz and Gomide, FSE 2007
-
Gradual fuzzy models
Ri: The more X is Ai the more Z is Ci
i = 1,..., N
IIN
i
N
ii
iii
ii CCC
xAzCyxR
11)(
otherwise0)()(if1),(
=
=
==
=
Pedrycz and Gomide, FSE 2007
-
Gradual fuzzy model architecture
Poss 1 C1
Min
A1 C1
Poss i
Ai Ci
Poss N
AN CN
C x
Ci
CN
C1
C1
C1
Pedrycz and Gomide, FSE 2007
-
x z
A1 A2
2
1
C1 C2 1 1
1
2 C1 C2
Example: gradual fuzzy model processing
Pedrycz and Gomide, FSE 2007
-
ExampleP: X is x inputs x [0, 3]
R1: The more X is A1 the more Z is C1rules
R2: The more X is A1 the more Z is C1
1 1.5 2 2.5 3 3.5 40
0.2
0.4
0.6
0.8
1
x
A
i
A1 A2
1 1.5 2 2.5 3 3.5 40
0.2
0.4
0.6
0.8
1
z
C
i
C1 C2
1 1.5 2 2.5 3 3.5 41
1.5
2
2.5
3
3.5
4
x
z
output
Pedrycz and Gomide, FSE 2007
-
11.6 Approximation propertiesof fuzzy rule-based models
Pedrycz and Gomide, FSE 2007
-
FRBS uniformly approximates continuous functions any degree of accuracy closed and bounded sets
Universal approximation with (Wang & Mendel, 1992): algebraic product t-norm in antecedent rule semantics via algebraic product rule aggregation via ordinary sum Gaussian membership functions sup-min compositional rule of inference pointwise inputs centroid defuzzification
Pedrycz and Gomide, FSE 2007
-
Universal approximation when (Kosko, 1992): min t-norm in antecedent rule aggregation via ordinary sum symmetric consequent membership functions sup-min compositional rule of inference pointwise inputs centroid defuzzification
(additive models)
Pedrycz and Gomide, FSE 2007
-
Universal approximation with (Castro, 1995): arbitrary t-norm in antecedent rule semantics: r-implications or conjunctions triangular or trapezoidal membership functions sup-min compositional rule of inference pointwise inputs centroid defuzzification
Pedrycz and Gomide, FSE 2007
-
11.7 Development of rule-basedsystems
Pedrycz and Gomide, FSE 2007
-
Expert-based development
Knowledge provided by domain experts basic concepts and variables links between concepts and variables to form rules
Reflects existing knowledge can be readily quantified short development time
Pedrycz and Gomide, FSE 2007
-
Example: fuzzy control
FuzzyController Processr
e u y+
Ri: If Error is Ai and Change of Error is Bi then Control is Ci
Ri: If e is Ai and de is Bi then u is Ci
Pedrycz and Gomide, FSE 2007
-
Ri: If e is Ai and de is Bi then u is Ci
Change of Error (de) / Error (e) NM NS ZE PS PMNB PM NB NB NB NM
NM PM NB NS NM NM
NS PM NS Z NS NM
Z PM NS Z NS NM
PS PM PS Z NS NM
PM PM PM PS PM NM
PB PM PM PM PM NM
r
y
t
e
Pedrycz and Gomide, FSE 2007
-
Data-driven development
Given a finite set of input/output pairs{(xk, yk), k = 1,..., M}xk = [x1k, x2k,...., xnk] Rn
zk = [xk, yk] Rn+1, k = 1,..., M
Clustering zk = [xk, yk] Rn+1, k = 1,..., M (e.g. using FCM)v1, v2,....,vN prototypes/cluster centersvi Rn+1, i = 1,..., N
Idea: fuzzy clusters fuzzy rules
Pedrycz and Gomide, FSE 2007
-
Example
R1v1
v2v3
v4
R2R3
R4
Pedrycz and Gomide, FSE 2007
-
Projecting the prototypes in the input and output spaces
v1[y], v2[y],....,vN [y] projections of prototypes in Y
v1[x], v2[x],....,vN [x] projections of prototypes in X
Ri: If X is Ai then Y is Ci, i = 1,..., N
x
y
x
y
x
y
Pedrycz and Gomide, FSE 2007
-
11.8 Parameter estimation forfunctional rule-basedsystems
Pedrycz and Gomide, FSE 2007
-
Functional fuzzy rules
Ri: If Xi1 is Ai1 and ... and Xin is Ain then z = aio + ai1x1 + ....+ ainxni = 1,..., N
Given input/output data: {(x1, y1), (x2, y2),....,(xM, yM)}
Let ai = [aio, ai1, ai2,...., ain]T
Output of functional models
Output for linear consequents
=
=
== N
iki
kiik
N
iikiikk
x
xw,,fwy
11 )(
)()( ax
TT
1
T ]1[ kikikN
iiikk w,,y xzaz ==
=
Pedrycz and Gomide, FSE 2007
-
[ ]
=
=
N
Nkkkk
N
y
a
a
a
zzz
a
a
a
aM
LM
21
TT2
T1
21
=
=
TT2
T1
T2
T22
T12
T1
T12
T11
21
NMMM
N
N
M
Z
y
yy
zzz
zzz
zzz
y
L
MOMM
L
L
M
Let
and
then y = Za
Pedrycz and Gomide, FSE 2007
-
Global least squares approach
Mina JG(a) = || y Za||2
|| y Za||2 = (y Za)T (y Za)Solution
aopt = Z# y
Z# = (ZT)1ZT
Pedrycz and Gomide, FSE 2007
-
Local least squares approach
Solution
aiopt = Zi# y
Zi# = (ZiT)1ZiT
=
= =
T
T2
T1
1
2)(JMin
iM
i
i
i
N
iiiL
Z
Z
z
z
z
ayaa
M
Pedrycz and Gomide, FSE 2007
-
11.9 Design issues of FRBS:Consistency andcompleteness
Pedrycz and Gomide, FSE 2007
-
Given input/output data: {(x1, y1), (x2, y2),....,(xM, yM)}
dataconsistencycompletenessaccuracy
rules
Issue: quality of the rules
Pedrycz and Gomide, FSE 2007
-
Completeness of rules
All data points represented through some fuzzy setmaxi = 1,..., M Ai(xk) > 0 for all k = 1,2,..., M
Input space completely covered by fuzzy setsmaxi = 1,..., M Ai(xk) > for all k = 1,2,..., M
Pedrycz and Gomide, FSE 2007
-
Consistency of rules
Rules in conflict similar or same conditions completely different conclusions
Conditions and Conclusions
Similar Conclusions
Different Conclusions
Similar Conditions rules are redundant rules are in conflict
Different Conditions
different rules; could be eventually
mergeddifferent rules
Pedrycz and Gomide, FSE 2007
-
|})()(||)()({|)cons(1
kjkikjM
iki xAxAyByBj,i =
=
Ri: If X is Ai then Y is BiRj: If X is Aj then Y is Bj
is an implication induced by some t-norm (r-implication)
))}()(Poss())()({Poss()cons(1
kjkikjM
iki yB,yBxA,xAj,i =
=
Alternatively
=
=
M
jj,i
Mi
1)cons(1)cons(
Pedrycz and Gomide, FSE 2007
-
11.10 The curse of dimensionalityin rule-based systems
Pedrycz and Gomide, FSE 2007
-
Curse of dimensionality number of variables increase exponential growth of the number of rules
Example n variables each granulated using p fuzzy sets number of different rules = pn
Scalability challenges
Pedrycz and Gomide, FSE 2007
-
11.11 Development scheme offuzzy rule-based models
Pedrycz and Gomide, FSE 2007
-
Accuracy Stability
Interpretability Knowledge
Representation
Spiral model of development incremental design, implementation and testing multidimensional space of fundamental characteristics
Pedrycz and Gomide, FSE 2007