Information Technology Reference
In-Depth Information
10
8
6
4
2
0
−2
−4
−6
−8
−10
−10
−8
−6
−4
−2
0
2
4
6
8
10
x
Fig. 12.11
Approximation of the test function of Eq. (
12.25
)(
dashed line
), using a feed-forward
neural network with Hermite basis functions
10
8
6
4
2
0
−2
−4
−6
−8
−10
−10
−8
−6
−4
−2
0
2
4
6
8
10
iterations
Fig. 12.12
Approximation of the test function of Eq. (
12.25
)(
dashed line
), using a one hidden
layer FNN with sigmoidal basis functions
f.x/D 20:5e
.0:3jxj/
sin.0:03x/ cos.0:7x/
(12.25)
4. Test function 4 given in Eq. (
12.26
) over the domain D D Œ10;10 Œ10;10:
f.x;y/D sin.0:3x/ sin.0:3y/e
.0:1jyj/
(12.26)
The approximation results for the 2D function of Eq. (
12.26
), obtained by an
NN with Hermite basis functions are shown in Fig.
12.15
. The approximation