PLI31 C Genetikoi Algorithmoi & Efarmoges

E
T N E
T '
A
E
A E
T N E
'
A E
YPIN YKOANAH
A T M H/Y

ATPA 2001

HPOOPIKH

T N E
'
A E
YPIN YKOANAH
A T M H/Y

KPATH KATIKA
T M A

EPIO NIKHOPIH
T I

ANATAIO TPIANTH

TEANO OYNTZH

...

TYPORAMA

/ 2000
ISBN: 9605381753
K : H 31/3
Copyright 2000

& , 26222 : (0610) 314094, 314206 : (0610) 317244
. 2121/1993,
.

............................................................................................................................................................. 9
K 1
, , ,
E ................................................................................................................. 13
1.1
A
, ,
E ................................................................................................. 16
1.1.1 .............................................................. 17
1.2
A
, ................................................................... 20
1.2.1 A .................................... 25
1.2.2 ......................................................................................................... 27
............................................................................................................................................................... 29
B
................................................................................................................................................
31
K 2
, , ,
E ................................................................................................................. 32
2.1
2.2
2.3
.................................
35
.................................................
37
............................................................
48
............................................................................................................................................................... 55
B
................................................................................................................................................
58
ENETIKOI AOPIMOI KAI EAPMOE
K 3
A A
, , ,
E ................................................................................................................. 59
3.1
A A
, ................................................................... 61
3.2
, ................................................................... 79
............................................................................................................................................................... 83
B
................................................................................................................................................
85
K 4
, , ,
E ................................................................................................................. 87
4.1
A
, ................................................................... 89
4.2
Y ....................................................................................... 102
............................................................................................................................................................ 106
B
.............................................................................................................................................
108
K 5
A A
, , ,
E .............................................................................................................. 109
5.1
A
, ,
E ............................................................................................... 111
5.1.1 K GRAY ................................................................... 111
5.2

, ,
E .............................................................................................. 115
5.2.1 ............................................................................................... 115
EPIEXOMENA
5.2.2 ........................................................... 116

5.2.3 ............................................................................................................... 117
5.2.4 ............................................................................................................ 119
5.2.5 ......................................................................................................... 122
5.2.6 ............................................................................................................ 122
............................................................................................................................................................ 125
B
.............................................................................................................................................
126
K 6
E A
, , ,
E .............................................................................................................. 127
6.1
T M
, ................................................................ 129
6.2
A
T N
, ................................................................ 133
6.3

, ................................................................ 142
............................................................................................................................................................ 152
B
.............................................................................................................................................
154
A A A ...................................................................... 155
Y A
.....................................................................
171
......................................................................................................................................................
183

, , ,
(). , ,
1998. ,
.
, ,
.
,
.

. , ,
, , . ,

(Artificial Intelligence)
(Machine Learning),
(Computational Intelligence).
, ( ). ( ) ( ) ( ) .
: , .
, ,
.
.
10
.

.
,
. , , ,
.

.
( ),
. , .

.
() ().
,
, , ,
.
.
A,
. ,
, . ,

.

A. ,
, , . -
POOO
, . ,
, . , . , ,

.
( , , )
, ,
.. , , . , , , . A .
A. ,
.
. , ,

. A . ,
.
.
:
, . . . , .
, -
11
12

. , , , , ,
, ,
.., .
, ,
, . , ()
, .
, ,
.
,
, .
, ,
. .
, , , ,
.
, ,
.

, .
,
,

. .
,
(A). .
,
. A, .
,
.

, :
A
.
.
A.
A.
, A.

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY
14

,
. ,
,
, :
(calculusbased methods):
. :
.
.

(hillclimbing).
. . ,
.
(enumerative) (random) :
.
( ) ,
.
, ,
. , .
(iterated search):
.
hillclimbing ( ),

.
.
( ), -
E I A I K E A PAT H P H E I
, .
(Simulated Annealing): hillclimbing. , , .
(Dynamic Programming): .
, .
.
(heuristic methods): ,
.
,
. ,
. , .
,

. , . ,
. , , ,
.
,
,
.
15
16
1.1 A
, , , , ,
.
, .

,
( , ,
) .

,

.
, , ' .
, .
. , ,
, ,
.
,
1.1 A
,

.
, .

1950,
. , ,
,
1970 John Holland
Michigan [2].
1.1.1
(Evolution of Species)
, ,
. ,
, ,
. , . ,
, :

( , ..
). ,
. ,
,
. , ,
, .
17
18
, ,
(chromosomes), .
,
(genes).
(genotype).
.
,
, (phenotype).

(reproduction) (mutation).
, (..
), . , , , ,
.
,

. , ,
. ,
,
, .. , , .. . , (..
) ,
. (dominant) (recessive). ( ),
(alleles).
John Holland, ,
'70 [2]. Holland
1.1 A
19
, ,
. olland ,

.
(A)
. , ,
.

. ,
.

,
. , , . ,
,
,
, . , ,
,
. ,
, ,
.
, , , .
1.1
1.2
20
1.3
1.1
, ;
.
1.2 A
1.1.1
. , .
.
.

A . ,
,
A
, A
. ,
.
.
, , (
46 ). A
.
. -
1.2 A
.
(loci). ( ) ,
.
, , ( ).
( ) .

, .

.

, .
A A . , , ,
, A
. ,
. , A ,
. , A [3].
, , ,
. A .
,
-
21
22
,
.
A .
, .
A
.
. ,
, .
(objective fitness
function), . .
:
t, A :
P (t ) = { x1t ,K, x nt } .
xit
.
, ( t
+ 1) . (mutation) (crossover mating)
.
.
,
( a1 , b1 , c1 , d1 , e1 ) ( a2 , b2 , c2 , d2 , e2 ) , (
crossover point = 2) ( a1 , b1 , c2 , d2 , e2 )
1.2 A
( a2 , b2 , c1 , d1 , e1 ) . . . ,
, . , ,
, . , ,
.

.
(mutation rate).

, , .
A
:
1. .
2. .
3. , .
4. .
5. ( , , ..).
. , [, ] , .
, ,
.
23
24
1.4
1.5
,
, ,
,
.
A , :
1. E
.
2. E .
3. , .
4. T 1 2.
5. T 1 3.
6. .
.
1.6
1.2
()
f(x) = x2 x [0, 31]. x 0 31;

. (
),
.
. ,
, .. 0, 1, 16, 31 . , , k, . , .
1.2 A
1.2.1 A
A
. , , :
1. .
A . , , ,
A. , A
.
2. . A
,
.
, . ,

,
.
3. . , A ,
, . , A,
, . , .
4. .
A , , , ,
A .
A
5. .
, ,

25
26
. , A , , ,
.
6. .
, , , . A .
7. . A . . A , ,
.
8. . A
,
.
O (n 3 ) , 10 1000. , ,
.
9. . .
,
. ,
(hillclimbing)
, .
. A
,
.
10. . A
,
1.2 A
27
, . , .
A :
1. .
2. ,
,
.
3. .
4. , .
5. .
.
1.2.2
A ,
.
.
, ,
;
, :
1. .
,
, .
, ' . , A .
A , , -
1.7
28

.

, , .
, ,
A , , .
(.. : , , .). , A . ,
,
. ,
A , .
2. .
, , . , ,

. , , :
,
, ; . ' , ,
.

, . ,
.
, . , ,
, , .
, ,

1.2 A
. , ,
. , ,
,
. .
, , , , . ,
, A . , , (. ) .
, .
. . ,
.
. , . ,
,
. ,
, .
,
.

.
,
, -
29
30
. , . , A , . T, A, E
.
. , .
A ,
.
.
, A.
, A ,
. , .
.
A
, . ,
,
A.
, .
1,
.
:
A
,
.
A.
A.
, A.
B I B I O PA I A
, () .
;
[1] Goldberg D.E.,GENETIC ALGORITHMS in Search, Optimization and

Machine Learning, Addison Wesley Publishing Company, Inc., 1989.
[2] Holland J.H., Adaptation in Natural and Artificial Systems, M.I.T. Press,
1975.
[3] Michalewich Z., Genetic Algorithms + Data Structures = Evolution
Programs, SpringerVerlag, 2nd ed., 1992.
[4] Mitchel, Melanie, An Introduction to Genetic Algorithms, MIT Press,
1996.
[5] Davis L., Handbook of Genetic Algorithms, Van Nostrand Reinhold, 1991.
31
1.8
, , ,
.
A .

, ,
, , . ,
, . :

(, , )

, , .

KEAAIO 2: A
34

,
. ,
, , .
A .
A,
. ,
. .
.
, .
( , ,
), .
2 . 1 A
2.1 A
A . , . , , D.
Goldberg [1], :
1. A
: O A , ,
. , :
(onoff). s
f ( s ) .
. ,

,
. A, ,
.
, . 1 on 0 off. , 11110

on off.
. , , .
, ,
A .
2. A :
,
,
. , , ,

, . A

( ).
35
KEAAIO 2: A
36
(hillclimbing) ,
.
,
, , ( ).
, A
. , .. 01101, 11000, 01000 10011.
, ,
. .
3. A : .
A
,
. ,
. ' A (KnowledgeBased Genetic Algorithms).
4. A : H
A,
. ,
. , ,
.

A , 1.2.
2.1
A ;
.
2 . 1 A
.
1 2 16 .
3, 4 5 6, 7
8 .
.
2.2 A
, A , .
.
, ,
.
A
. A

. ,
, . ,
()

. , ( ).
, .
. (bits):
(binary string) . , ,
. , -
37
2.1
KEAAIO 2: A
38
. A , , .
. , ,
( ) .
:

; .
, .
2.1
f(x) = x2, x [0, 31] x: . .
A . , ,
5, .
[0, 31] 32 . ..
10010 , , 18
. , , .

A
. ( ),
.
.

. ,
.
, .
2.2 A
2.1, f,
.
, , x, (fitness score),
, f.
,
,
A. .

.
, ,
.
A, .

A.
. A
:
1. (Initialization)
2. (Decoding)
3. (Fitness calculation
evaluation)
4. (Reproduction)
I.
(Selection)
II. (Crossover mating)

III. (Mutation)
5. (2) A
39
40
KEAAIO 2: A
,
A.
, (, , ).
,
, .
2.1, 4.
32 .
20 , 4 5 .
01101, 11000, 01000 10011.
, A
. ,
.
. , :
. ,
, . . ,
, .
.

. ,
. , , , ,
.

.
. , , . , ,
, .. -
2.2 A
,
. ,
A
( ):

. ,
o.

A,
. . . : . ,
.
, ,
. ,

.
,
, . A .
A.
, ,
, ,
, , .
, .
, ,
, 1 [1].
, ,
20 . , -
41
KEAAIO 2: A
42
( ),
, 2.1:
2.1
1.
01101
169
14.4
2.
11000
576
49.2
3.
01000
64
5.5
4.
10011
361
30.9
1170
100.0
1170.
.

2.1.
A3
5%
A2
50%
A4
31%
2.1
A1
14%
, .
, 1 169,
14.4% .
1 14,4%
, 0,144.
,
() ,
. ,
2.2 A
, , ,
. [1] (mating pool).

, . , ,

. , , .
.
, .

A. ,
.
, .

. A , A,
, [3].
.

. ,
. , , ,
, . , -
43
KEAAIO 2: A
44
, , (crossover probability) pc,

A , ,
. , , . pc=1, ,
. , , . , p c
, . X ,
, .
, ,
, , . M , ,
.

, ,
, , .

.
. :
.
, , (mutation probability) pm, (
0 1 ). (
), A .
,
, . , , ,
. , , -
2.2 A
45
.
, , P(t)
t, 2.2.
Procedure Genetic Algorithm
begin
t0
P(t)
P(t)
while ( not ) do
begin
t t+1
P(t) P(t1)
P(t)
P(t)
end
end
2.2
,
, .
f(x) = x2 [1, 31], 2.2,

( = 1
= 0).
, ,
.
(. 2.2 )
.
pm = 0.002, 20 (4 * 5) bits ;
2.2
KEAAIO 2: A
46
2.2
T x
(
)
f ( x ) = x2
pselecti =
fi
A/
fi
f
01101
11000
01000
10011
f
f /4
Maximum
2.2
2.3
2.2, , ;
string :
k 1 l
string [1, l1]. strings k+1 l .
=112=3 =012
=1, strings;
) 1: 0 1 1 0 1
2: 1 1 0 0 0
2.4
) 1: 1 1 0 0 0
2: 1 0 0 1 1
A
2.2, pc = 1 pm = 0,001. (2.3).
2.3
2.2 A
47
2.3
Z
( )

( )
T x
(
)
01101
11000
11000
10011
f
f /4
Maximum
A .
.
, , .
, ,
.
: 5, 10, 15, 25, 50, 100. , ,

n = 6.
f ( x ) = x2
2.3
2.4
2.5
KEAAIO 2: A
48
2.6

,
, ncounti : ncounti = f i / f , f i ( ) i f .
, i string ,

. , ncounti 1,25,
i 1,0
0,25. , 6 . , .
2.3
A
,
.
2.2
:
f ( x ) = x sin(10px ) + 1.0
x [1, 2] f, x0
f ( x0 ) f ( x ) , x [1, 2].
H f .
f
f '( x ) = sin(10px ) + 10px cos(10px ) = 0 ,
tan(10px ) = -10px
[3] ,

2.3
xi =
2i - 1
+ ei , i = 1, 2, ,
20
x0 = 0 ,
2i + 1
- ei , i = 1, 2, ,
20
ei
i = 1, 2, i = 1, 2, .
xi =
f ,
xi i
xi i .
[1, 2], x19 = 37 / 20 + e19 = 1.85 + e19 , f(x19)

p
f (1.85 ) = 1.85 sin(18p + ) + 1.0 = 2.85 .
2
, A .
, A 2.2, f.
,
.
x. , . x 3. [1, 2] 3 1000000 . , 22
,
2097152 = 221 < 3000000 222 = 4194304
<b21b20 b0> x [1, 2]
:
:
49
KEAAIO 2: A
50
( < b21b20 ...b0 > )2 = (
21
i =0
bi 2i )10 = x '
x
x = -1.0 + x '
3
,
2 -1
22
1.0 3
. ,
(1000101110110101000111)
0.637107,
x' = (1000101110110101000111)2 = 2288967
x = -1.0 + 2288967
3
= 0.637197
4194303
,
(0000000000000000000000) (1111111111111111111111)
, 1 2 .

. , 22 . 22 .

eval v
f:
eval ( v ) = f ( x ) ,
v x.
,
,
. ,
v1 = (1000101110110101000111),
v2 = (0000001110000000010000 ),
v 3 = (1110000000111111000101),
2.3
x1 = 0.637197, x2 = 0.958973 x3 = 1.627888,

. , :
eval ( v1 ) = f ( x1 ) = 1.586345,
eval ( v2 ) = f ( x2 ) = 0.078878,
eval ( v 3 ) = f ( x 3 ) = 2.250650.
, v3 ,
.

A
.
.
, . v3
. 0 1 v3
v 3 ' = (1110100000111111000101).

x 3 ' = 1.721638 f ( x 3 ' ) = -0.082257.

v3. ,
v3 ,
v 3 '' = (1110000001111111000101).

x 3 '' = 1.630818 f ( x 3 '' ) = 2.343555.

v3,
f ( x 3 ) = 2.250650.

v2 v3. , ,
. T
51
KEAAIO 2: A
52
v2 = (00000 | 01110000000010000 ),
v 3 = (11100 | 00000111111000101)
:
v2 ' = (00000 | 00000111111000101),
v 3 ' = (11100 | 01110000000010000 ).

f ( v2 ' ) = f ( -0.998113 ) = 0.940865,
f ( v 3 ' ) = f (1.666028 ) = 2.459245.

.
2.2, A:
pop _ size = 50
pc = 0.25
pm = 0.01

2.4,
A .
2.4
1.441942
2.250003
2.250283
2.250284
10
2.250363
12
2.328077
2.3
53
39
2.344251
40
2.345087
51
2.738930
99
2.849246
137
2.850217
145
2.850227
A,
,
. 150 :
v max = (1111001101000100000101),
x max = 1.850773.
x max = 1.85 + e19 f ( x max )
2.85.
2.4 pc 1,
2.2 p c = 0.25.
(
) ; .
1. , .
, . , ;
1: 1 1 0 1 | 1 0 0 1 0 1 | 1 0 1 1
2: 0 0 0 1 | 0 1 1 0 1 1 | 1 1 0 0
2. , : .
,
, .
2.7
2.5
KEAAIO 2: A
54
, ,
,
.
. ;
1: 1 0 0 1 0 1 1
2: 0 1 0 1 1 0 1
: 1 1 0 1 0 0 1
2.6
2.8
2.9
,
100 (
100), (1).
;

, . X
, , ,
1000 0 1. : 0.00 0.25, 0.25 0.50, 0.50 0.75, 0.75 1.00
. ; ;
10 : 0.1, 0.2, 0.05, 0.15, 0.11, 0.07, 0.04, 0.12, 0.16.

,
. , 10 . 1000
,
.
2.3

,
. strings, 10: 1011101011, 0000110100.
: 3, 1, 6 20.
,
pm.
1000 ,
p m = 0.001, 0.01, 0.1.
.

. :
1. .
2. .
3. , .
4. ,
.

, , .
.
, .
,
.
, , .
55
2.10
2.11
KEAAIO 2: A
56
, .
. ,
A .
A ,
.
. , .

. , A
:
1. (Initialization)
2. (Decoding)
3. (Fitness calculation evaluation)
4. (Reproduction)
I. (Selection)
II. (Crossover mating)
III. (Mutation)
5. (2) A
, ,

/ ,
.
,
. , . ,
A ,
YNOH
. ,
A.
.
,
.
. , ,
A.
. A,
. ,
.
A
,
. ,
, A. , .
2, , .

,
. :
,
,
,
( , , ).
, , .
57
KEAAIO 2: A
58
2.7
() .


1975.
1996.
A A

. , A,
.

A. , ,
A.
,
. , . .
,
.
,
.

, :
, A,
, /, A,
,
,

A, ,

.
KEAAIO 3: A A
60

[1]:
,
. ,
. , A , ,
, , .
, .

, ,
. ,
.
,
, .
, ,
A,
. , -
3.1 A A
,
, , (),
A .
, , A.
. . ,
,
.
, 2. , ,
.
3.1 A A

. , , A. ,
. ,

. , , , . , , .

, :
61
62
KEAAIO 3: A A
, A,
,
A, /,

,
,
,

A, .
, A
.

f(x), x = xmax,
f (xmax) = max. f. f, g, g = f.
, f
, C,

max g ( x ) = max{ f ( x ) + C } .
, , k , f ( x1 ,K, x k ): R k R . xi Di = [ ai , bi ] R f ( x1 ,K, x k ) > 0, "xi Di , i = 1,K, k . f , ..
q .
,
:
3.1 A A
63
1. .
2. .
3. .
4. .
5. .
6. .
7. () .
,
.

.

q , Di = [ ai , bi ] ( bi - ai ) 10 q
. mi
( bi - ai ) 10 q 2mi - 1 . T,
mi q i .
, bin _ str , :
xi = ai + decimal ( bin _ str )
bi - ai
2mi - 1 ,
(3.1)
decimal ( bin _ str )

bin _ str .
1.2. ' ,
m=
k
i =1
mi ,
(3.2)
KEAAIO 3: A A
64
m1 x1,
[ ai , bi ] , m2 x2 [ a2 , b2 ] , ...
.

. ,
.
.
A . ,
( ), ( ) ( ), .
3.1
2.2 2.3 2.
.
. ,
().
3.1
x1
x2
...
F(x1,x2,,xl)
01111
15
00111
...
225
...
01001
00010
...
101
...
00111
01001
...
123
...
00111
00101
...
81
...
3.1 A A
,
, .
, Pascal,
[1]:
, , pop_size
m.
.
(type) (type declaration). , population array
individual ( 1 pop_size). individual
record chrom,
chromosome, fitness
x. , x, . , chromosome array
allele ( 1 m),
boolean ( bit
true false). .

.
, ,

.
, pop_size m
. :
1. , (
f ).
2. , .
3. .
65
KEAAIO 3: A A
66
4. ,
, 3.1
.
5. , , A .
( ).
T
1
2
3
N1
N
T+1
1
2
3
N1
N
3.1

A.
, (slotted roulette wheel).

:
1.
eval ( vi )
vi , i = 1,K, pop _ size .

2.
F=
pop _ size
i =1
eval ( vi ) .
3. pi
vi , i = 1,K, pop _ size : pi = eval ( vi ) / F .
4. , (cumulative) qi
vi , i = 1,K, pop _ size : qi =
i
j =1
pj .
3.1 A A

pop_size . , :
1. r 0 1.
2. r < q1, v1, vi ( 2 i pop _ size ), qi -1 < r qi .
, .
, .
A pc.
:
1. r 0 1.
2. r < pc, .
(
pc pop_size),
pos [1,
m1], m .
pos . ,
:
( b1b2 K bpos bpos +1 K bm )
( c1c2 K c pos c pos +1 K cm )

( b1b2 K bpos c pos +1 K cm )
( c1c2 K c pos bpos +1 K bm ) .
, , .
. A, pm.
pm m
pop_size. :
67
KEAAIO 3: A A
68
:
1. r 0 1.
2. r < pm, .
,
. ,
. .
3.1
3.2
pos . ;
50 33,
, 0.001, 0.01 0.1;
[3],
.
3.1
A .
pop_size = 20, pc = 0.25 pm = 0.01.
:
f ( x1 , x2 ) = 21.5 + x1 sin( 4px1 ) + x2 sin( 20px2 ) ,
3.0 x1 12.1 4.1 x2 5.8.
.
x1 15.1, [3.0,
12.1] 15.1 10000 = 151 000 -
3.1 A A
. 18 x1 ( ), :
217 < 151000 218 .
x2 15
. , , (3.2),
m = 18 + 15 = 33 .
:
(010001001011010000111110010100010 ) .
18 ( 010001001011010000 )
( 3.1)
x1 = -3.0 + 70352
15.1
= -3.0 + 4.05242 = 1.05242 ,
262143
15 ( 11110010100010 )
x2 = 5.75533 . , (x1, x2) =
(1.05242, 5.75533).
f (1.05242, 5.75533) = 20.25264.
:
v1 = (100110100000001111111010011011111)
v2 = (111000100100110111001010100011010 )
v 3 = (000010000011001000001010111011101)
v 4 = (100011000101101001111000001110010 )
v5 = (000111011001010011010111111000101)
v6 = (000101000010010101001010111111011)
v7 = (001000100000110101111011011111011)
v8 = (100001100001110100010110101100111)
v9 = (010000000101100010110000001111100 )
v10 = (000001111000110000011010000111011)
69
70
KEAAIO 3: A A
v11 = (011001111110110101100001101111000 )
v12 = (110100010111101101000101010000000 )
v13 = (111011111010001000110000001000110 )
v14 = (010010011000001010100111100101001)
v15 = (111011101101110000100011111011110 )
v16 = (110011110000011111100001101001011)
v17 = (011010111111001111010001101111101)
v18 = (011101000000001110100111110101101)
v19 = (000101010011111111110000110001100 )
v20 = (101110010110011110011000101111110 )
. :
eval(v1) = f(6.084492, 5.652242) = 26.019600
eval(v2) = f(10.348434, 4.380264) = 7.580015
eval(v3) = f(2.516603, 4.390381) = 19.526329
eval(v4) = f(5.278638, 5.593460) = 17.406725
eval(v5) = f(1.255173, 4.734458) = 25.341160
eval(v6) = f(1.811725, 4.391937) = 18.100417
eval(v7) = f(0.991471, 5.680258) = 16.020812
eval(v8) = f(4.910618, 4.703018) = 17.959701
eval(v9) = f(0.795406, 5.381472) = 16.127799
eval(v10) = f(2.554851, 4.793707) = 21.278435
eval(v11) = f(3.130078, 4.996097) = 23.410669
eval(v12) = f(9.356179, 4.239457) = 15.011619
eval(v13) = f(11.134646, 5.378671) = 27.316702
eval(v14) = f(1.335944, 5.151378) = 19.876294
eval(v15) = f(11.089025, 5.054515) = 30.060205
eval(v16) = f(9.211598, 4.993762) = 23.867227
eval(v17) = f(3.367514, 4.571343) = 13.696165
3.1 A A
71
eval(v18) = f(3.843020, 5.158226) = 15.414128

eval(v19) = f(1.746635, 5.395584) = 20.095903
eval(v20) = f(7.935998, 4.757338) = 13.666916
v15
v2 .
(roulette wheel).
(fitness) :
20
F=
eval ( v ) = 387.776822
i
i =1
pi vi , i=1,,20, :
p1 = eval(v1)/F = 0.067099
p11 = eval(v11)/F = 0.060372
p2 = eval(v2)/F = 0.019547
p12 = eval(v12)/F = 0.038712
p3 = eval(v3)/F = 0.050355
p13 = eval(v13)/F = 0.070444
p4 = eval(v4)/F = 0.044889
p14 = eval(v14)/F = 0.051257
p5 = eval(v5)/F = 0.065350
p15 = eval(v15)/F = 0.077519
p6 = eval(v6)/F = 0.046677
p16 = eval(v16)/F = 0.061549
p7 = eval(v7)/F = 0.041315
p17 = eval(v17)/F = 0.035320
p8 = eval(v8)/F = 0.046315
p18 = eval(v18)/F = 0.039750
p9 = eval(v9)/F = 0.041590
p19 = eval(v19)/F = 0.051823
p10 = eval(v10)/F = 0.054873 p20 = eval(v20)/F = 0.035244

(cumulative probabilities) qi
vi ,i=1,,20 :
q1 = 0.067099
q6 = 0.293917
q11 = 0.538381
q16 = 0.837863
q2 = 0.086647
q7 = 0.335232
q12 = 0.577093
q17 = 0.873182
q3 = 0.137001
q8 = 0.381546
q13 = 0.647537
q18 = 0.912932
q4 = 0.181890
q9 = 0.423137
q14 = 0.698794
q19 = 0.964756
q5 = 0.247240
q10 = 0.478009
q15 = 0.776314
q20 = 1.000000
20 ; . 20 [0, 1]:
KEAAIO 3: A A
72
0.513870
0.175741
0.308652
0.534534
0.947628
0.171736
0.702231
0.226431
0.494773
0.424720
0.703899
0.389647
0.277226
0.368071
0.983437
0.005398
0.765682
0.646473
0.767139
0.780237
r = 0.513870 q10
q11, v11
. r = 0.175741
q3 q4, v4 . :
v1* = (011001111110110101100001101111000) (v11)
v2* = (100011000101101001111000001110010) (v4)
v3* = (001000100000110101111011011111011) (v7)
v4* = (011001111110110101100001101111000) (v11)
v5* = (000101010011111111110000110001100) (v19)
v6* = (100011000101101001111000001110010) (v4)
v7* = (111011101101110000100011111011110) (v15)
v8* = (000111011001010011010111111000101) (v5)
v9* = (011001111110110101100001101111000) (v11)
v10* = (000010000011001000001010111011101) (v3)
v11* = (111011101101110000100011111011110) (v15)
v12* = (010000000101100010110000001111100) (v9)
v13* = (000101000010010101001010111111011) (v6)
v14* = (100001100001110100010110101100111) (v8)
v15* = (101110010110011110011000101111110) (v20)
v16* = (100110100000001111111010011011111) (v1)
v17* = (000001111000110000011010000111011) (v10)
v18* = (111011111010001000110000001000110) (v13)
v19* = (111011101101110000100011111011110) (v15)
v20* = (110011110000011111100001101001011) (v16)
, ,
() . ,
3.1 A A
73
,
. , .
r 0 1.
pc = 0.25, r < 0.25, o .
:
0.822951
0.911720
0.031523
0.581893
0.151932
0.519760
0.869921
0.389248
0.625477
0.401154
0.166525
0.200232
0.314685
0.606758
0.674520
0.355635
0.346901
0.785402
0.758400
0.826927
v2, v11, v13 v18

( ,
,
).
, . (v2 v11) (v13 v18).
, , pos = 9:
v2 = (100011000101101001111000001110010 )
v11 = (111011101101110000100011111011110 ) .

v2 = (100011000101110000100011111011110 )
v11
= (111011101101101001111000001110010 ) .
, , pos = 20:
v13 = (000101000010010101001010111111011)
v18 = (111011111010001000110000001000110 ) .

v13
= (00010100001001010100 0000001000110 )
v18
= (111011111010001000111010111111011) .
74
KEAAIO 3: A A
:
v1 = (100110100000001111111010011011111)
v2 = (100011000101110000100011111011110 )
v 3 = (000010000011001000001010111011101)
v 4 = (100011000101101001111000001110010 )
v5 = (000111011001010011010111111000101)
v6 = (000101000010010101001010111111011)
v7 = (001000100000110101111011011111011)
v8 = (100001100001110100010110101100111)
v9 = (010000000101100010110000001111100 )
v10 = (000001111000110000011010000111011)
v11
= (111011101101101001111000001110010 )
v12 = (110100010111101101000101010000000 )
v13
= (00010100001001010100 0000001000110 )
v14 = (010010011000001010100111100101001)
v15 = (111011101101110000100011111011110 )
v16 = (110011110000011111100001101001011)
v17 = (011010111111001111010001101111101)
v18
= (111011111010001000111010111111011)
v19 = (000101010011111111110000110001100 )
v20 = (101110010110011110011000101111110 )
, , pm = 0.01 ,
1% . m pop_size = 33 20 = 660
, 6.6 .
r [0, 1]. r < 0.01, .
, 660 . ,
, ,
3.1 A A
, , :
112
349
418
429
602
4
11
13
13
19
13
19
22
33
8
, . ,
:
v1 = (100110100000001111111010011011111)
v2 = (100011000101110000100011111011110 )
v 3 = (000010000011001000001010111011101)
v 4 = (011001111110010101100001101111000 )
v5 = (000111011001010011010111111000101)
v6 = (000101000010010101001010111111011)
v7 = (001000100000110101111011011111011)
v8 = (100001100001110100010110101100111)
v9 = (010000000101100010110000001111100 )
v10 = (000001111000110000011010000111011)
v11
= (111011101101101001011000001110010 )
v12 = (110100010111101101000101010000000 )
v13
= (000101000010010101000100001000111)
v14 = (010010011000001010100111100101001)
v15 = (111011101101110000100011111011110 )
v16 = (110011110000011111100001101001011)
v17 = (011010111111001111010001101111101)
v18
= (111011111010001000111010111111011)
75
76
KEAAIO 3: A A
v19
= (111011100101110000100011111011110 )
v20 = (101110010110011110011000101111110 )
.
, :
eval ( v1 ) = f (3.130078, 4.996097 ) = 23.410669
eval ( v2 ) = f (5.279042, 5.054515 ) = 18.201083
eval ( v 3 ) = f ( -0.991471, 5.680258 ) = 16.020812
eval ( v 4 ) = f (3.128235, 4.996097 ) = 23.412613
eval ( v5 ) = f ( -1.746635, 5.395584 ) = 20.095903
eval ( v6 ) = f (5.278638, 5.593460 ) = 17.406725
eval ( v7 ) = f (11.089025, 5.054515 ) = 30.060205
eval ( v8 ) = f ( -1.255173, 4.734458 ) = 25.341160
eval ( v9 ) = f (3.130078, 4.996097 ) = 23.410669
eval ( v10 ) = f ( -2.516603, 4.390381) = 19.526329
eval ( v11 ) = f (11.088621, 4.743434 ) = 33.351874
eval ( v12 ) = f (0.795406, 5.381472 ) = 16.127799
eval ( v13 ) = f ( -1.811725, 4.209937 ) = 22.692462
eval ( v14 ) = f ( 4.910618, 4.703018 ) = 17.959701
eval ( v15 ) = f (7.935998, 4.757338 ) = 13.666916
eval ( v16 ) = f (6.084492, 5.652242 ) = 26.019600
eval ( v17 ) = f ( -2.554851, 4.793707 ) = 21.278435
eval ( v18 ) = f (11.134646, 5.666976 ) = 27.591064
eval ( v19 ) = f (11.059532, 5.054515 ) = 27.608441
eval ( v20 ) = f (9.211598, 4.993762 ) = 23.867227
3.1 A A

. F
447.049688,
387.776822. , , v11, , v15, 33.351874 30.060205.
, :
;
:
,
. , A , . , 100 1000 0.001%.
:
1.
,
.
. .
2. . (
)
. ,

( ).
77
KEAAIO 3: A A
78
3.3
3.1 2 ,
, .
3.4
3.1,
;
3.5
3.6
3.7
3.1
3.1
; 0.25 20, ; ,
; .
,
,
;
( ),
;
3.1,
. .
3.2
, 1.
100 ,
, 1000 .
79
3.2
1.
.
2. 50
1.
3. 1 2.
3.1 100
. 0.1 1.0, 0.1.
,
. .
( 2.5) 3.1.
,
3.2, 100 .
3.2

A, ,
, .
A, .
,
3.3
3.4
80
KEAAIO 3: A A
(redundant values). ,
2.

, .

:
.
.
,
3.1, A.

A. , A . ,
,
. ,
,
.
. A,
.
A
. , ,

, . A,
.
, 2. .
3.2
81

: ( ). ,
,
(.. g(x)
= 0 g(x) 0
g(x) 0. , .
, , .
, , (
' )
.
,
,

.

. , , A.
.
A (penalty method).
: ( ). , :
E g(x) hi(x) 0, i = 1, 2, , n,
x m.
:
n
E g ( x ) + r
F[h ( x )] , i
i =1
r .
r A
82
KEAAIO 3: A A

(redundant values). ,
2. , A. ..,
[0 9]
,
4 0000 1111. , ,
1010 1111;
;
,
A, . , , ( ) , [3]:
1. .
2. ,
.
3. .
, ,
, .
:
1. (fixed remapping),
, .. 10 - 15
0 - 5 . ,
.
2. (random remapping),
.
, A -
YNOH
83
g(x) h(x) 0 y(x) 0. ;
f(x,y,z) .
x [20.0, 125.0], y
[0, 1.2 106] z [0.1, 1.0].
( )
0.5, 104 0.001 . .
; , , : (20, 0, 1), 125.0, 1.26, 1.0) (50, 100000, 0.597).
;
A , , ,
, . , ,
,
.
, ,
3.8
3.9
3.5
84
KEAAIO 3: A A
.
,
,
.
,
,
A, . ,
,
, , (),
A . , . ,
, ,
.
, .
A, , , .
A, . ,
(redundant
values). ,
2. 3.2,
.
, A. 3,
A , Pascal. , 3.1. , , A .
3,
YNOH
, .
, ,
. ,
,
., . ,
.
B

85

. ,
,
A.

. ,

A ( ) . (building blocks) ( ), A.
,
.

, :
,
,

,
,
A.

KEAAIO 4: A
88

A ,

. A, [1, 2, 4].
.
A
,
.
[4]:
A; ,
( )
A;
(, ) A;
A ;
A ;
A , ;
A;
4.1 A
( A ) A
;
, . .
, [6, 7, 8].
4.1 A
, .
,
.
;

.
. ,
( ), A.

, :
,
,
.
89
90
KEAAIO 4: A
, .
. A, Holland [2], ,
, A
, () .
, A , .
()
. Holland ,
.
()
A ,
(schema) (template) .
(don't care symbol) *
( = {0, 1}). ( ), *.
,
10. (*111100100) :
{(0111100100 ), (1111100100 )}
(*1*1100100) :
{(0101100100 ), (0111100100 ), (1101100100 ), (1111100100 )} .
(1001110001) ,
(1001110001), (**********) 10. 2r , r * .
, m 2m
4.1 A
. D. Goldberg [1, .19]

. m
3m , *. ,
(cardinality) c (c + 1)m. ,
(1001110001). 210 :
(1001110001)
(*001110001)
(1 * 01110001)
(10 *1110001)
M
(100111000*)
(** 01110001)
(*0 *1110001)
M
(10011100 **)
(***1110001)
M
(**********) .
.
*
.
:
(order) (defining length). (Schema Theorem) .
(SCHEMA THEOREM)
S ( (S))
0 1, (fixed positions), *. ,
*.
91
KEAAIO 4: A
92
(speciality) ,
. , , 10,
S1 = (***001 *110 ) ,
S2 = (**** 00 **0*) ,
S 3 = (11101 **001) ,
:
o( S1 ) = 6 , o( S2 ) = 3 o( S 3 ) = 8 ,
S3 , , , , S1
S2 16 128 .
.
S ( (S)) .
(compactness) .
,
d ( S1 ) = 10 - 4 = 6 , d ( S2 ) = 9 - 5 = 4 d ( S1 ) = 10 - 4 = 6 .
,
.
.
A
( ).

.

.
pop_size = 20
( ) m = 33
4.1 A
( 3.1). , (
) t :
v1 = (100110100000001111111010011011111)
v2 = (111000100100110111001010100011010 )
v 3 = (000010000011001000001010111011101)
v 4 = (100011000101101001111000001110010 )
v5 = (000111011001010011010111111000101)
v6 = (000101000010010101001010111111011)
v7 = (001000100000110101111011011111011)
v8 = (100001100001110100010110101100111)
v9 = (010000000101100010110000001111100 )
v10 = (000001111000110000011010000111011)
v11 = (011001111110110101100001101111000 )
v12 = (110100010111101101000101010000000 )
v13 = (111011111010001000110000001000110 )
v14 = (010010011000001010100111100101001)
v15 = (111011101101110000100011111011110 )
v16 = (110011110000011111100001101001011)
v17 = (011010111111001111010001101111101)
v18 = (011101000000001110100111110101101)
v19 = (000101010011111111110000110001100 )
v20 = (101110010110011110011000101111110 )
(S, t) t
S. ,
S0 = (****111 **************************) ,
(S0, t) = 3, ( v13, v15 v16),
S0. S0 o(S0) = 3 (S0) = 7 5 = 2.
93
94
KEAAIO 4: A
t eval (S,
t).
t S. p { vi1 ,K, vi p } t
S. ,
p
eval ( S , t ) = (
eval ( v
ij
)) / p .
(4.1)
j =1
, .
, , . ,
, vi
p1 = eval (vi)/F(t), F(t)
.
, (S, t + 1) S.
1. S, eval (S, t)/F(t),
2. S (S,
t)
3. pop_size,

x ( S , t + 1) = x ( S , t ) pop _ size eval ( S , t ) / F (t )
(4.2)

F (t ) = F (t ) / pop _ size , :
x ( S , t + 1) = x ( S , t ) eval ( S , t ) / F (t ) .
(4.3)
, , .

4.1 A
95
. ,
.
:
S
% ( eval ( S , t ) = F (t ) + e F (t ) ),
x ( S , t ) = x ( S , 0 ) (1 + e )t ,
e = (eval ( S , t ) - F (t )) / F (t )
(4.4)
> 0 < 0
.
. ,
, .
S0.
t S0,

eval ( S0 , t ) = ( 27.316702 + 30.060205 + 23.867227 ) / 3 = 27.081378 .
,
20
F (t ) = (
eval ( v )) / pop _ size = 387.776822 / 20 = 19.388841

i
i =1
S0
eval ( S0 , t ) / F (t ) = 1.396751 .
S0
. , t
S0 1.396751 ,
t + 1 3 1.396751 = 4.19 ( 4 5), t + 2 3 1.3967512
= 5.85 ( 5 6), ...
, S0 -
96
KEAAIO 4: A
, ,
.
. t S0 . ,
:
v1 = (011001111110110101100001101111000 )
( v11 )
v2 = (100011000101101001111000001110010 )
( v4 )
v 3 = (001000100000110101111011011111011)
( v7 )
v 4 = (011001111110110101100001101111000 )
( v11 )
v5 = (000101010011111111110000110001100 )
( v19 )
v6 = (100011000101101001111000001110010 )
( v4 )
v7 = (111011101101110000100011111011110 )
( v15 )
v8 = (000111011001010011010111111000101)
( v5 )
v9 = (011001111110110101100001101111000 )
( v11 )
v10
= (000010000011001000001010111011101)
( v3 )
v11
= (111011101101110000100011111011110 )
( v15 )
v12
= (010000000101100010110000001111100 )
( v9 )
v13
= (000101000010010101001010111111011)
( v6 )
v14
= (100001100001110100010110101100111)
( v8 )
v15
= (111001100110000101000100010100001)
( v20 )
v16
= (111001100110000101000100010100001)
( v1 )
v17
= (111001100110000100000101010111011)
( v10 )
v18
= (111011111010001000110000001000110 )
( v13 )
v19
= (111011101101110000100011111011110 )
( v15 )
v20
= (110011110000011111100001101001011)
( v16 )
, S0 :
v7 , v11
, v18
, v19
v20
.
, ,
4.1 A
( ) . ,
.

, ,
.
, . ,
.
, ..
v18
= (111011111010001000110000001000110 ) ,
233 . ,
:
S0 = (****111 **************************)
S1 = (111 ****************************10 ) .

( 3). , , pos = 20. S0
, S0.
111 ,
, ..
v18
= (111011111010001000110000001000110 )
v13
= (00010100001001010100 0000001000110 )

v18
= (111011111010001000111010111111011)
v13
= (00010100001001010100 0000001000110 ) .
, S1 ,
. 111 10 .
,
-
97
98
KEAAIO 4: A
. , S0
(S0) = 2, S1 (S1) = 32.
, (uniformly) m
1 . S
pd ( S ) =
d (S )
m -1
(4.5)
:
ps ( S ) = 1 -
d (S )
.
m -1
(4.6)
, S0 S1
pd ( S0 ) = 2 / 32 , p s ( S0 ) = 30 / 32 , pd ( S1 ) = 32 / 32 , p s ( S1 ) = 0 ,
.

, pc . , :
p s ( S ) = 1 - pc
d (S )
.
m -1
(4.7)
, (pc = 0.25)
p s ( S0 ) = 1 - 0.25
2
= 63 / 64 = 0.984375 .
32
,
, . , v18
v13
111 10, S1 . ,
p s ( S ) 1 - pc
d (S )
.
m -1
(4.8)
4.1 A
99
d (S )
x ( S , t + 1) x ( S , t )eval ( S , t ) / F (t )1 - pc
m - 1
(4.9)

, . ,
.
S0:
d ( S0 )
eval ( S0 , t ) / F (t )1 - pc
= 1.396751 0.984375 = 1.374927 .
m - 1
, S0

. t + 1, 3 1.374927 = 4.12
t + 2, 3 1.3749272 = 5.67 .

pm.

. , ,
v19
= (111011101101110000100011111011110 )

S0 = (****111 * ** ********* **************) .
, , v19
.
3.1 3, v19

:
v19
= (111011100101110000100011111011110 ) ,
S0. 1 4 8 33,
S0. 3 ( , ) : -
KEAAIO 4: A
100
. ,
, .
pm, 1 pm.
, ( )
p s ( S ) = (1 - pm ) o( S )
(4.10)
, , Pm << 1,
p s ( S ) 1 - o( S ) pm .
(4.11)
S0
pm = 0.01
p s ( S0 ) 1 - 3 0.01 = 0.97 .
, ,
d (S )
x ( S , t + 1) x ( S , t ) eval ( S , t ) / F (t )1 - pc
- o( S ) pm
m -1
(4.12)

( ) .
S0
d (S )
eval ( S0 , t ) / F (t )1 - pc
- o( S ) pm =
m
1
= 1.396751 - 0.954375 = 1.333024.

, S0 ( , ,
, )
: t + 1 3 1.333024 = 4.00 S0,
t + 2 3 1.3330242 = 5.33 .

4.1 A
101
(4.12) ( ):
(SCHEMA THEOREM)
,

.
m
2m .
> 0,
.
< 0.
f
x, l = 4,
4.1
4.2
4.3
x (.. f 0011 = 3, f 1111 = 15 ).

1***, f; 0*** f;
x,
x.
S, k
1,
l k.
4.4
102
KEAAIO 4: A
4.2
A ,
,
. ():
: , ,
, .

, :
S0 = (****111 * ** ********* **************) .
, ,
( ) .
.
,
. , A, . ,
A
.
pop_size
m, 2m 2pop_size . :
() (
).

A. , , ,
. ,
, 11
4.2
S1 = ( 1 1 1 * * * * * * * * )
S2 = ( * * * * * * * * * 1 1 )
( ),
S3 = ( 1 1 1 * * * * * * 1 1)
, , :
S4 = ( 0 0 0 * * * * * * 0 0)
S0 = ( 1 1 1 1 1 1 1 1
1 1 1) ( S3 ). A S0 ( 0 0 0 1 1 1 1 1 1 0 0 ).
: ( , )
A .
[3].

( ). ,
, , , , .
, , . , ( ) :
,
.
: , . ,
s = ((1, 0), (2, 0), (3, 0) | (4, 1), (5, 1), (6, 0), (7, 1) | (8, 0), (9, 0), (10, 0), (11, 1))
,
s = ((1, 0), (2, 0), (3, 0) | (7, 1), (6, 0), (5, 1), (4, 1) | (8, 0), (9, 0), (10, 0), (11, 1))
, , -
103
KEAAIO 4: A
104
.
S3 = ( 1 1 1 * * * * * * 1 1)

S3 = ((1, 1), (2, 1), (3, 1), (4, *), (5, *), (6, *), (7, *), (8, *), (9, *), (10, 1), (11, 1)),

S3 = ((1, 1), (2, 1), (3, 1), (11, 1), (10, 1), (9, *), (8, *), (7, *), (6, *), (5, *), (4, *)),
.
.
, [3]: (messy) .
, [4].
.
, , ,
5 [3].
4.1
A1 = 11101111, A2 = 00010100
A3 = 01000011 H1 = 1*******, H2 = 0*******, H3
= ******11, H4 = ***0*00*, H5 = 1*****1* H6 = 1110**1*.
) ;
) .
) ,
pm = 0.001.
) , pc = 0.85.
4.2
105
0 :
A/A
10001
20
11100
10
00011
01110
15
pm= 0.01
pc = 1.0.
) S1=(1****)
1.
) S2=(0**1*)
1.
4.2
KEAAIO 4: A
106

,
. , , ,
. ,
.
() , .
,
. ,
, ,
. A .
( , )
,
, .
A. , A .
,

, , .
, A,
.
, A,
.
.
A, .
YNOH
, , .
:
,
,

,
,
A.
107
KEAAIO 4: A
108

1975.
1996.
[5] Whitley, L.D., Foundations of Genetic Algorithms 2, Ed., Morgan
Kaufman, 1993.
[6] Whitley, L.D. and Vose M.D., Foundations of Genetic Algorithms 3, Ed.,
Morgan Kaufman, 1995.
A A
,
. , ,
.
() . ,
.
, ,
.
, ,
.

, :
,
,
,

,

.

110
KEAAIO 5: A A

A,
. , .
A , () .
A
, , . , 100 [500, 500]
, 3000.
, ,
101000. A .

.

A.

. ,
() . , ,
.
5 . 1 A
5.1 A

.
(
) A
.

, :
,
.

,
, A
.
, .
,

. ,
. ,

Gray.
5.1.1 K GRAY
b = <b1,...,bm>
Gray g = <g1,...,gm> . m
.
111
KEAAIO 5: A A
112
M A Gray
begin
g1=b1
for k=2 to m do
gk=bk1 XOR bk
end
Gray
Gray
M A Gray
begin
value=g1
b1=value
for k=2 to m do
begin
if gk=1 then value=NOT value
bk=value
end
end
5.1 16
Gray.
Gray

. ,

. ,
Gray . ( m = 4),
1
1
A=
0
0
1
1
0
0
0
1
1
0
,
0
1
A -1 =
1
:
g=Ab b=A1g,
2.
0
1
1
1
0
0
1
1
0
0
5 . 1 A
113
, ,

.
,
.
. ,
.
Gray
0000
0000
0001
0001
0010
0011
0011
0010
0100
0110
0101
0111
0110
0101
0111
0100
1000
1100
1001
1101
uk, k = 0, 1, , N1,
1010
1111
1011
1110
1100
1010
1101
1011
1110
1001
1111
1000
N -1
J = min x N2 +
x k2 + uk2
k =0

x k +1 = x k + uk , k = 0, 1,..., N 1,
x0 , xk R
u RN . :
J * = K0 x02 ,
Kk Riccati, k = N1, , 0
K k = 1 + K k +1 / 1 + K k +1 K N = 1.
u.
(200, 200) ui ( ).
x0 = 100 N = 45,
u = u0 ,..., u44 , J* = 16180.4.
5.1
Gray
KEAAIO 5: A A
114
5.1
Gray
:
.
.
. .
.
.
. .
5.2
:
. .
. .
. .
. .
. .
5.1
, .
5.2
5.2

( ) .
. , .
.

:

,
,
, ,
.

,
1.
. .
: . ,
.
5.2.1
115
116
KEAAIO 5: A A
,
(
,
) .
,
.
, ,
,

.
( )
(UB LB) / (2n 1), UB LB
n
.
5.2.2

, .

.
,
.
, ,
.
,
( ). ,

,
. , -
5.2

.
5.2.3
DEC3100
[1]. 10
. 60
20000.
n = 30 ( ), 30 45 = 1350
.
,

.
. , .

( ) .

. ,
,
.
0.25, (
2).
117
KEAAIO 5: A A
118
5.2
0.6
0.7
0.8
0.9
0.95
0.00047
0.00068
0.00098
0.0015
0.0021
0.014
0.02
0.03
0.045
0.061

(
) (0.25). ,
,
.
(LB, UB).
5.3

0.6
0.7
0.8
0.9
0.95
42179
46102
29290
52769
30573
31212
46594
41806
47454
69624
82371
11275
5.3
:
u1 = [0.1, 0.3, 1.2, 0.05]
u2 = [1.5, 0.05, 0.8, 0.3]
,
.
5.2
5.2.4
3
. , ,
(16180.4). , ,

, .
,
.

, . :
% ( 400, 200, -200 )
; :
:
d , 2d .
, = 0.05 0.05, 0.1 .
:
. n = 30 m
, m
m n + log 2 d .
m m = n + log 2 d = 25
m/n = 25/30 = 0.833, . , , .

,
, ,
119
KEAAIO 5: A A
120
.

.

t
: su = u1 ,..., um -
( t ) k
, sut +1 = u1 ,..., u k ,..., um ,
(
(
u k + D t ,UB -u k ), 0,
u k =
u k - D t ,u k - LB ), 1,
LB UB
k. (t, y) [0, y]
(t, y) , t .
( t
) . ,
,
.
b
t
1-
D t , y = y 1 - r T ,
( )
r [0, 1], b
(
b = 5).

, ,
. , u k
5.2
121
( )
u k = u k , t , n) ,
n = 30 . (k, i) :
k i ( 0 )
( )
( )
t,n 0,
t , n =
t,n

1,
b ,
( b = 1.5).
( )
[*]

3.
5.4

0.8
0.9
35265
30373
40256
20561
26164
2133
( 4). ,
. , ,
, (16205 16189).
5.3.
[*]
To
.
5.4
KEAAIO 5: A A
122
5.2.5
, . , .

. ,
, .
.
5.2.6
.
, .
5.5

0.7
0.8
0.9
23814
19234
27456
6078
16188.2
16248
16798
16198
54
16182.1
5.2
123
A
. .

5.2.5.
5.6
( )
.
A (N)
15
25
35
45
1080
3123
5137
7177
9221
184
398
611
823
1072
CPU
.

30 bits .
, .
5.7
( )
, =45.

10
20
30
40
50
4426
5355
7438
9219
10981
12734
1072 ()
KEAAIO 5: A A
124
5.2
2.097.152 . A A .
, :
1. ,
2. ,
3. ,
4.
.
5.3
5.4
A A
. ,
A , f(x) = x10,
A , . 30 10 . .
YNOH
,
. ,
, .
.
,
. , .
.
, ( ) (
,
).
( ).
, , ,
.
.
,
. ,
. ,
.
, , .
125
126
KEAAIO 5: A A
, .

. , .
B

E A
,

, . ,
,
.

. ,

. ,
, .

. , :
A ,
, A,
A, ,
,
.

KEAAIO 6: E A
128

(blueprint)
(markers)

, A . , ,

() . A
(Image Processing),
/ (Computer Aided Desing CAD), , , T (Software Engineering), (Scheduling), (Computer Graphics)
.

A
. A,
, .
.
6.1 T M
6.1 T M
, A
, .

, :
A ,
,
A,
.

. ,
,
,
.
.
, , , .
A , . , ,
, , ..
2%, .
, , .
129
KEAAIO 6: E A
130
, ,
, , .
,
. , A [1], [2], [3].

(structural optimization). Goldberg [2]
A . 10
, ,

. A :
, . . 4 10
,
. , ,
.
, A
. , A

, .
,
Powel, Skolnik Tong [3],
.
, ,
: (), () A. ,
6.1 T M

.
, ,
,
.
(Interdigitation) .
,
. , A
. .
,
: ,
. , . , A. , , A .
.
, .
.
, ,
.

.
.
, , . General Electric .
General Electric, ( ), A Engineous. -
131
132
KEAAIO 6: E A
, ,
CAD. Engineous,

[4]. ,
A. ,
, , , . ,
Engineous
,
(trial and error). A
100 , 10.
6.1
:
. ,
. ,
. ,
. .
6.2
,
, ;
6 . 2 A T N
6.2 A
T N
, ,
. A .

, :
,
,

,
. . ,
, , , ,
, , , .
A, ,
,
. A
()
, ( A) .
[3].
133
134
KEAAIO 6: E A
,
(),
. ,
. ,
.
, ,
(SigmaPi unit),
,
. ,

,
.
(product
unit). M ,
,
N
y=
X (i )
p( i )
i =1
p(i)
.
,
,
, , . ,
,
.
.
(Product Neural Networks) (). 6.1.

6 . 2 A T N
135
6.1:
X1
X2
X3
X4
(Error Back Propagation BP).

, . , ,
, ,
. ,
.
, A,
. , Janson Frenzel [3], A,
.
A ' :
, .
. , 30
100, 32 . 37 , 1184 .
: (AT)
KEAAIO 6: E A
136

1
.
1 + ATL
1.
. , :
20 , 20 .
. ,
1000 .

CAD CMOS. ,

CMOS, , .
, ,
. , A
. ,
. , A 5 20
.

( A) . '
, ,
A
.
. ,
A -
6 . 2 A T N
, .

.

Harp Samad [3]. A
.
,
. , , ,
. ,
. , , , '
, .
, Harp Samad [3]
, NeuroGENESYS, A
.
. , ( )
. , . NeuroGENESYS , , , ..
A
. , ,
,
137
138
KEAAIO 6: E A
, A
, , .
A
, , , ..
,
, , .. .
() . , , , , (learning rate),

(feedback connections). ,

.
, A .
, , . , , , .. ( blueprint ),

.
,
( ) ( ). :
1. ()
2. ().
, , , .. ,
6 . 2 A T N
, ..
, ,
.
, A. , ,
A .

(markers)
. ,
.
,
A. 30 100
. ,
.
. .
,
,
.
:
1. , .
2. E
.
0.01 .
, ,
, , . -
139
140
KEAAIO 6: E A
( ). :
1. ,
,

2. , .
, ( ). , ,
(.. , , .).
F(i) i:
n
F (i ) =
a Y ( p (i )) .
j
j =1
, pj ,
j. NeuroGENESYS
j .
NeuroGENESYS .
,
ExclusiveOR.
, .
. (fanout) ,
fanout
. , ,
fanout.
ExclusiveOR (EXOR), NeuroGENESYS (, , ).
6 . 2 A T N
141
, , 6.2.
INPUT
OUTPUT
6.2:

NeuroGENESIS
, 0.75, 0.25.
, Harp
Samad [3].
Jansen & Frenzel

.
6.3
Jansen &
Frenzel .
6.4

Harp & Samad,
;
Harp & Samad

n
F (i ) =
a Y ( p (i ))
j
j =1
pj aj
F.
6.5
6.1
KEAAIO 6: E A
142
6.3
A
,
. ,
.
. ,
, .

,
:
A , , ,
A,
,
.
(scheduling)
,
[5].
, , (NPcomplete) .
, .

,
6.3
, , . . (),
,
. A
.

(System Integration Test Station)
. ,
F14, (cockpits), (radar), ., . , , , , , . .
( ), . ,
,
1. , ..
.
2. , ..
5 ..
3. , .. , ,
.
Gilber Syswerda [3]
(scheduling system) ,
. A
. , -
143
144
KEAAIO 6: E A
. , .
, (deterministic
schedule builder),
. FCFS (First Come First Served),
,
, , ... (
). .
,
. A,
.
.
, , , , '
. ,
,
.

A ,

.

. , .
, .
,
.

-
6.3
.
, .

,
. ,
,
,
( ) .
, , ,
A
. , . , ,
,
, ,
. , .
, :
(OrderBased Crossover).
(PositionBased Crossover).
(Edge Recombination Crossover).
. ()
.
.
1:
,
:
145
KEAAIO 6: E A
146
123456789
412876935
****
,
. ( ) 2, 8, 6 5.
2, 5 ,6 8. , ( 2865).
,
2, 5, 6, 8:
1
(1 x 3 4 x x 7 x 9).

, 2, 8, 6, 5,
1
(1 2 3 4 8 6 7 5 9 ).
2
:
1
abcdefghij
eibdfajgch
****
:
1
aibcfdeghj
ibcdefahjg
Whitley [3],
.
.
:
(Positionased utation).
(OrderBased Mutation).
6.3
(Scramble mutation).

.
,
. ,

.
.

, .
Whitley, Starkweather Shaner [3]
GENITOR,
A.
(), . .
, . ..
ABCDEF : A AB, BC, CD, DE, EF
FA. ,
AB BA .

.
. (edge map)
.

( ).

(Robotics) .
(robot trajectory generation) -
147
148
KEAAIO 6: E A
, ,
(rules) .
,
. ,
, ,
.
' ,
,
, A, .
A , ,
A .
O Yuval Davidor [6],
, A
.
A,
.
.
,
(endeffecter)

.
( ) ,
,
.
,
6.3
149
, . ,

.
,
, . A Davidor [6]
, . .

. :
n - tuple
644
744
8
a1,1; a1,2 ;L; a1, n ; a2 ,1;L; a2 , n ;L; al , n
14444442444444
3
l n - tuples
,
n
n () . , l ,
ln.
.
A .
.

(Analogous Crossover) [6].
,
. , ,
. 6.3.
150
KEAAIO 6: E A
6.3
M 1
M 2
, '
( A) . ,
, . ,
. .
,
. ,

, ,
. ,
(Segregation
Crossover), :
1. .
2. n .
3.
n n. n
.
4. 1 2
.
5.
.
.
:
6.3
151
(Addition Mutation):
n .
.
(Deletion Mutation): n,
,
.
, A
. 4.2 6.3.
1040. 100. A
: .
.

. A
.
A , .
A Holland
[7],
. , Prediction Company Santa Fe,
(
Prophet), A .
Man Machine Interfaces Inc.,
A
.
,
, 1 6.3.
6.7
KEAAIO 6: E A
152
6.2
, .
.
A
. , , . . , A
.
. ,
, A ,
.
A ( A),
.
A , A . A , ' .
, A
. [5]
A. A, .
A
. A
, (highly parallel algorithms).
YNOH
,

.
A.
A ,
. , A .
, A ,
. ,

A .

.
. ,

. .
A
. , ,
. ,
, ,
. , .
.
A
, . . , .
, , -
153
KEAAIO 6: E A
154
. :
A ,
, A,
A ,
,
.
B

[4] S. Hedberg, Emerging Genetic Algorithms, in Artificial Intelligence
Expert, September 1994.
[5] R. Mangano, A Genetic Algorithm White Paper, in An Introduction to
Genetic Algorithm Implementation: Theory, Application, History and
Future Pontential, Man Machine Interfaces Inc., 1993.
[6] Y. Davidor, A Genetic Algorithm Applied to Robot Trajectory Generation,
Proc. Of 3rd International Conference on Genetic Algorithms, 1989.
1975.

1.1
, :

.

(chromosomes), ,
.

(reproduction) (mutation).
,
,
.
1.2
, ,
.
.
, 1.1.1.
1.3
,
, ,
.. .
1.4
,
, ,

, . , , . , -
156
. ,
,
.
.
1.5
5. , 1.1.2.
1.6
(string) l = 5,
( ) 0 (00000) 31 (11111).
, . , 53 095, :
5 * 10 4 + 3 * 10 3 + 0 * 102 + 9 * 101 + 5 * 100 = 5395
, 0 1
10011 :
1 * 2 4 + 0 * 2 3 + 0 * 22 + 1 * 21 + 1 * 20 = 16 + 2 + 1 = 19
,
.
1.7
1.2
.
5.
1.8
, .
( )
. , .
, .
A A N T H E I A K H E N AY T O A I O O H H
157
2.1
. , , . , ,
. , , . ,
....
2.2
1.1.3, , :
1. .

T x
(
)
f ( x ) = x2
pselecti =
A/
fi
f
fi
01101
13
169
0.14
0.58
11000
24
576
0.49
1.97
01000
64
0.06
0.22
10011
19
361
0.31
1.23
1170
1.00
4.00
4.0
293
0.25
1.00
1.0
576
0.49
1.97
2.0
f
f /4
Maximum
2. bits : 4
5 0.001 = 20 0.001 = 0.02 < 1. , bit .
158
2.3
k = 3, k + 1 = 4
2.
) 0 1 1 0 1
01100
) 0 1 0 0 0
1 1 0 0 0
01011
11001
1 0 0 1 1
10000
2.4
, , . 2.2, . ,
.

Z
( )

( )
T x
(
)
f ( x ) = x2
01101
01100
12
144
11000
11001
25
625
11000
11011
27
729
10011
10000
16
256
1754
439
729
f
f /4
Maximum
2.5
1. , :
1: 1 1 0 1 0 1 1 0 1 1 1 0 1 1
2: 0 0 0 1 1 0 0 1 0 1 1 1 0 0

.
2. ,
:
1: 1 0 0 1 1 0 1
2: 0 1 0 1 0 1 1
, ,
,
,
, .
2.6
x , f(x) = x,
.
2.7

:
1. .
2. .
3. , .
4. , .
159
160
3.1
l , k
1 l1, pos = k [1, l]. .
3.2

pop_size m pm. :
1. 50 33 0.001 = 1.65
2. 50 33 0.01 = 16.5
3. 50 33 0.1 = 165
3.3
29 < 1510 < 210, m1 = 10 26 < 170
< 27 , m2 =7.
m = m1 + m2 = 10 + 7 = 17.
3.4
6.6
. 5 , .
3.5
20
0.25 = 5. 4 . , .
3.6
, ,
.
3.7
.
.
3.8
y'(x) = y(x), : y'(x) 0.
, :
g '( x ) = g ( x ) + r1F(h( x )) - r2 F( y ( x ))
3.9
:
1. .
2. ,
.
3. .
, ,
, . , :
1. (fixed remapping),
2. (random remapping)
4.1
, , .
, ( 4.1) .
, 10. (*111100100)
{(0111100100), (1111100100)}
161
162
(*1*1100100)
{(0101100100), (0111100100), (1101100100), (1111100100)}.
(1001110001) ,
(1001110001), (**********) 10. 2r , r * .
, m 2m
. , (1001110001).
210 :
(1001110001)
(*001110001)
(1 * 01110001)
(10 *1110001)
M
(100111000*)
(** 01110001)
(*0 *1110001)
M
(10011100 **)
(***1110001)
M
(**********) .
, m j = 1, , m ,
m j, :
m
=
j
j =0
m
m - j m
(1) (1)
j
j =0
163
( x + y ) = j ( x ) ( y )
m
m- j
j =0
, x = y = 1,
m
j = 2
j =0
4.2
S % ( eval ( S , t ) = F (t ) + e F (t ) ), :
x ( S , t ) = x ( S , 0 ) (1 + e )t
e = (eval ( S , t ) - F (t )) / F (t )
(4.4)
> 0 < 0
.
. ,
, . < 0, .
4.3
f
f = b0 20 + b1 21 + b2 22 + b3 2 3 , bi = 0 1 i = 0, 1, 2, 3.
S, p ,
p
( ) eval ( v ) / p ,
F S =
j =1
eval(vj) vj.
1*** :
1000 eval 1000 = 8
164
( )
1010 eval (1010) = 10
1011 eval (1011) = 11
1100 eval (1100) = 12
1101 eval (1101) = 13
1110 eval (1110) = 14
1111 eval (1111) = 15
1001 eval 1001 = 9
, S1 = (1***)
( ) (
( )
F S1 = 8 + 9 + 10 + 11 + 12 + 13 + 14 + 15 / 8 F S1 = 11.5
0*** :
( )
0001 eval (0001) = 1
0010 eval (0010) = 2
0011 eval (0011) = 3
0100 eval (0100) = 4
0101 eval (0101) = 5
0110 eval (0110) = 6
0111 eval (0111) = 7
0000 eval 0000 = 0
, S2 = (0***)
( ) (
( )
F S2 = 0 + 1 + 2 + 3 + 4 + 5 + 6 + 7 / 8 F S2 = 3.5
4.4
f :
f = b0 + b1 + + bl , bi = 0 1 i = 0,1,..., l .
S 2l k
165
k ( k l k )
l ( l ). , S :
2 l -k
( ) eval ( v ) / 2
F S =
l -k
j =1
= k + ( k + 1) + + l / 2l - k
( )
(
l (l + 1) k ( k - 1) l (l + 1) - k ( k - 1)
=
=
))
k + k + 1 + + l = 1 + 2 + + k + + l - 1 + 2 + + k - 1 =
2
()
F S =
( ) (
l l +1 - k k -1
2
l - k +1
5.1
. ,
1.
5.2
. ,
,
,
5.5.
5.3

30.
4 30 = 120 .

u1 = <100000000001000001100010111110,
011111111100111011011001110000, 100000001100010010011100010100,
100000000000000011010010011000>
166
u2 = <100000001111010111000011001111,
100000000000000011010010011000, 011111110111110011101101100100,
011111110111110011101110001111>
60 (
).
u1 = <100000000001000001100010111110,
011111111100111011011001110000, 011111110111110011101101100100,
011111110111110011101110001111>
u2 = <100000001111010111000011001111,
100000000000000011010010011000,
100000001100010010011100010100,
100000000000000011010010011000>.

u1 = [0.1, 0.3, 0.8, 0.3] u2 = [1.5, 0.05, 1.2, 0.05].
5 u1 10
u2. :
u1 = <100000000001000001100010111110,
011101111100111011011001110000, 011111110111110011101101100100,
011111110111110011101110001111>
u2 = <100000001111010111000011001111,
100000000000000011010010011000,
100000001000010010011100010100,
100000000000000011010010011000>.

u1 = [0.1, 12.8, 0.8, 0.3] u2=[1.5, 0.05, 0.8, 0.05].
:
.

u1 = [0.1, 0.05, 0.8, 0.3] u2 = [1.5, 0.3, 1.2, 0.05].
167
u1 u2.
, ,
3.4 4.9. T
:
u1 = [0.1, 3.4, 0.8, 0.3] u2 = [1.5, 0.3, 1.2, 4.9].
5.4

u1 u2.
( ))
u k = u k , t , n ,
( )
( )
t,n 0,
t , n =
t,n 1,
( )
b
t
1-
D t , y = y 1 - r T ,
( )
= 20000, t = 1, b = 5 r
[0, 1].
0, 19
u1, 20
u1.
0, 19 u2, 20
u2.
T
u1 = [0.09925, 0.3, 1.2, 0.05] u2 = [1.5, 0.05, 0.80022, 0.3],
0
u1 = [0.10039, 0.3, 1.2, 0.05] u2 = [1.5, 0.05, 0.79936, 0.3],
1.
168

u1 u2.
:
(
(
u k + D t ,UB -u k ), 0,
u k =
u k - D t ,u k - LB ), 1,
b
t
1-
D t , y = y 1 - r T ,
( )
= 20000, t = 1, b = 5 r
[0, 1].
T
u1=[0.1, 0.05, 131.709, 0.3] u2=[66.991, 0.3, 1.2, 0.05],
0
u1=[0.1, 0.05, 131.453, 0.3] u2=[66.481, 0.3, 1.2, 0.05],
1.
6.1
. , A .
6.2
, .
, . .
, A .
.
6.3
:
, -
..,
.
: (AT)

1
.
1 + ATL
. , .
.
, 1000
.
6.4
A
(Error Back Propagation BP).
, . , ,
, . ,
.
, A,
.
Janson Frenzel A,
.
6.5
. .
,
,
.
169
170
6.6
. , ,
, , . (
).
6.7

1 6.3, :
2
(3 1 2 8 7 4 6 9 5).

1.1
,
(.. ,
)
. . , , .
.
, . ,
.
,
. , .
, ,
.
1.2
,
. .
. ,
.
l ,
6 1.2. .
, .
2.1
.
[1].
172
2.2
6 7,
, .
5 7, . .
, .
2.3
,
. ,
, 729,
576. ,
. , :
,
1024, .
. , .
2.4
, select ,
. select, , . , partsum.
rand
,
rand : = random sumfitness,
random 0 1. ,
sumfitness ( ), random. , repeatuntil
, ,
Y O E I E I A A N T H E N PA T H P I O T H T N
rand. j, select.
. , .
, .
2.5
.
2.6
2.4, .
2.7
A .
. , /,
,
. ,
. pc
, pc = 1
. ,
. ,
f ( x ) = x 2 , ,
x.
(pc = 1),
x, . pc,
. , ,
, . ,
, .
173
174
2.8
. , Matlab, , rand.
2.9
, , ,
1.0.
1.0, 9 10. ,

.
.
2.4,
. , . ,
.
2.10

, GAlib.

() (singlepoint), ().
.
2.1.3. ,
. , . , 2.11, .
[1] ( 3, .
6365). ,
.
2.11
2.10.
,

. ,
.
3.1
,
:
1. .
2. .
3. .
4. .
5. .
6. .
7. () .

3.1. ,
. 2.
,
2.1.2.
3.2

3.1, . , ,
.
1. , . ,
. 100 , 1000.
175
176
, ( ) .
2. ( ),
.
3.3
, ,
,
. 1,
.
, ,
. ,
,
. , . .
3.4
,
. ,
2.5. . ,
, , .
A,
.
( 10 100), ( ) . ,
.
3.5
,
(3.1). (3.2).
177
, . .

.
4.1
) 1 1, 3, 5 6.
2 2.
3 2 3.
) S, o(S),
0 1. S (
(S))
. ,
(1) = 1
(1) = 11 = 0
(2) = 1
(2) = 11 = 0
(3) = 2
(3) = 87 = 1
(4) = 3
(4) = 74 = 3
(5) = 2
(5) = 71 = 6
(6) = 5
(6) = 71 = 6
) S pm
() (
p s S = 1 - pm
,
ps(H1) = 110.001 = 0.999
ps(H2) = 110.001 = 0.999
ps(H3) = 120.001 = 0.998
ps(H4) = 130.001 = 0.997
ps(H5) = 120.001 = 0.998
ps(H6) = 150.001 = 0.995
) ( ) 1 - o(S ) p
o S
178
S pc :
()
p s S 1 - pc
d S)
m -1
m .
,
ps(H1) = 10.850/7 = 1
ps(H2) = 10.850/7 = 1
ps(H3) = 10.851/7 = 0.878571
ps(H4) = 10.853/7 = 0.635714
ps(H5) = 10.856/7 = 0.271429
ps(H6) = 10.856/7 = 0.271429
4.2
, , ( )
x S , t + 1) x S , t ) eval ( S , t ) / F t )
(S, t + 1) , (S, t)
, eval(S, t) S
()
F t .
,
(S1) = 1 1 = 0
o(S1) = 1
(S2) = 4 1 = 3
o(S2) = 2
) eval (v ) / 4 = (20 + 10 + 5 + 15) / 4 = 12.5
F t =0 =
i =1
(
) (
)
eval ( S , t = 0) = (5 + 15) / 2 = 10
eval S1 , t = 0 = 20 + 10 / 2 = 15
2
179
, :
15
0
1 - 1 - 1 0.01 = 2.376
12.5
4
10
3
1 - 1 - 2 0.01 = 0.368
12.5
4
) x S1 , t = 1 = 2
) x S2 , t = 1 = 2
, , S1 S2 1.
5.1
, . .
, , .
5.2
,
, 3.
, ,
. .
221
27
2097152
2097152
3
4
21
221 50 221
27
27 50 27
5.3
A,

, f(x) = x10.
, ,
,
A.
180
, 3 .
.
6.1
15, pj j. j .
() , () fan out (fo). , , j,
:
F = 0.4 + 0.3 + 0.3 fo.
,
fan out.
6.2

. [1].
.
,

, , .
.
, A [1, 166179].

.
. -
. ,
, ,
. , , . , n , log2n .
nlog2n .
, , . ,
20 ( 5 ), 5 (..
10110) . .
' , ,
,
, .
( ) .
, , . , ' ,
,
, . ,
. .
v = i1i2 Kin i1 i2, ..,
in1 in in. <12n>.
(..
(greedy) ,
) <12n>.
181
182
: ,
.
.
.
, ,
, . , A
. ,
.
' ,
OX [1, . 26], : ,
. ,
< 1 2 3 4 5 6 7 8 9 10 11 12 > < 7 3 1 11 4 12 5 2 10 9 6 8 >

(4 5 6 7),

< 1 11 12 4 5 6 7 2 10 9 8 3 > .
,
. ,
, . A
, . A 100 , 20000 , , 9.4% .
: T *,
0 1.
: O ,

.
: H ..
, . , .
: H
.
: M ( ), . ,
, ( )
. .
: ()
() .
: H () t .
: .., CPU
.
: H ,
.
: T ,
..
: T .
.. .
( ): T
.
184
: M ( ) , () , ().

.
,
1, .
: T ,
.
: H , ,.
: H
. () , .
: M ,
(
) .
: M ,
. , , , . , , , .
: M ,
,
.
.
API
: H ( )
.

, , . .
: H , .
: .
: H
, , .
: H ( ) .
,
( 0 1
). ,
, ,
.
: O
.
: H .
: O , .
: . ...
: H
.
185
186
: H
, . O
.
: H .
.
: O ,
2. .
: H .
: .
,
.
() : (template),
*
( = {0, 1}).
( , ),

*.
: O ( ) 0 1, ,
*.
: ,

, . ,
.
API
(): M ,

. (),
. ,
.
: .
: T
,
.
: M ,
. .
, .
187

PLI31 C Genetikoi Algorithmoi &amp; Efarmoges

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PLI31 C Genetikoi Algorithmoi &amp; Efarmoges

Uploaded by

Copyright:

Available Formats

E

ENETIKOI AOPIMOI KAI EAPMOE

5.2.2 ........................................................... 116

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

KEAAIO 1: EIAH TOY ENETIKOY AOPIMOY

[1] Goldberg D.E.,GENETIC ALGORITHMS in Search, Optimization and

II. (Crossover mating)

, , (crossover probability) pc,

f(x) = x2 [1, 31], 2.2,

: 5, 10, 15, 25, 50, 100. , ,

( < b21b20 ...b0 > )2 = (

x1 = 0.637197, x2 = 0.958973 x3 = 1.627888,

10 : 0.1, 0.2, 0.05, 0.15, 0.11, 0.07, 0.04, 0.12, 0.16.

[1] Goldberg D.E.,GENETIC ALGORITHMS in Search, Optimization and

decimal ( bin _ str )

, (slotted roulette wheel).

vi , i = 1,K, pop _ size .

eval(v18) = f(3.843020, 5.158226) = 15.414128

p11 = eval(v11)/F = 0.060372

p12 = eval(v12)/F = 0.038712

p13 = eval(v13)/F = 0.070444

p14 = eval(v14)/F = 0.051257

p15 = eval(v15)/F = 0.077519

p16 = eval(v16)/F = 0.061549

p17 = eval(v17)/F = 0.035320

p18 = eval(v18)/F = 0.039750

p19 = eval(v19)/F = 0.051823

p10 = eval(v10)/F = 0.054873 p20 = eval(v20)/F = 0.035244

v2, v11, v13 v18

g(x) h(x) 0 y(x) 0. ;

[1] Goldberg D.E.,GENETIC ALGORITHMS in Search, Optimization and

. D. Goldberg [1, .19]

x ( S , t + 1) = x ( S , t ) pop _ size eval ( S , t ) / F (t )

eval ( v )) / pop _ size = 387.776822 / 20 = 19.388841

= 1.396751 - 0.954375 = 1.333024.

x (.. f 0011 = 3, f 1111 = 15 ).

[1] Goldberg D.E.,GENETIC ALGORITHMS in Search, Optimization and

m/n = 25/30 = 0.833, . , , .

[1] Michalewich Z., Genetic Algorithms + Data Structures = Evolution

(Error Back Propagation BP).

Jansen & Frenzel

Harp & Samad

[1] Michalewich Z., Genetic Algorithms + Data Structures = Evolution

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

1000 eval 1000 = 8

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

ENETIKOI AOPIMOI KAI EAPMOE

PLI31 C Genetikoi Algorithmoi & Efarmoges

PLI31 C Genetikoi Algorithmoi & Efarmoges