Design of Neural Networks - Gene Expression Programming - page 386

a.

01234567890123456 7890123456789012

26375471750

3456

DTQaababaabbaabba 86648 7682

W = {-1.64, -1.834, -0.295, 1.205, -0.807, 0.856, 1.702, -1.026, -0.417, -1.061}
T = {-1.14, 1.177, -1.179, -0.74, 0.393, 1.135, -0.625, 1.643, -0.029, -1.639}

m

m

D

-0.029

Q

T

-0.625

1.643

a a a a

b a

b

b.

01234567890123456 7890123456789012

2637547

3456

DTQaababaabbaabba 86648 1750 7682

W = {-1.64, -1.834, -0.295, 1.205, -0.807, 0.856, 1.702, -1.026, -0.417, -1.061}
T = {-1.14, 1.177, -1.179, -0.74, 0.393, 1.135, -0.625, 1.643, -0.029, -1.639}

d

d

D

0.393

T Q

-0.625 -0.625

a a a a a

b b

Figure 10.2. Illustration of Dw-specific transposition. a) The mother neural
network. b) The daughter NN created by transposition. Note that the network
architecture is the same for both mother and daughter and that W m = W d and
T m = T d . However, mother and daughter are different because different combina-
tions of weights and thresholds are expressed in these individuals.

W 0,0 = {-0.78, -0.521, -1.224, 1.891, 0.554, 1.237, -0.444, 0.472, 1.012, 0.679}
W 0,1 = {-1.553, 1.425, -1.606, -0.487, 1.255, -0.253, -1.91, 1.427, -0.103, -1.625}

0123456789012345601234567890123456
T aabbbabb291 39341QDbabbabb40396369 -[0]
Q Tababaab552 27879QDbabbaaa36972318-[1]

W 1,0 = {-0.148, 1.83, -0.503, -1.786, 0.313, -0.302, 0.768, -0.947, 1.487, 0.075}
W 1,1 = {-0.256, -0.026, 1.874, 1.488, -0.8, -0.804, 0.039, -0.957, 0.462, 1.677}

Note that the weights of the offspring are exactly the same as the weights of
the parents. However, due to recombination, the weights expressed in the