Information Technology Reference
In-Depth Information
0.8
0.6
0.4
0.2
0
−0.2
−0.4
−0.6
10
5
10
5
0
0
y
−5
−5
x
−10
−10
Fig. 12.19
Approximation of the test function of Eq. (
12.27
), using a feed-forward neural network
with Hermite basis functions
0.8
0.6
0.4
0.2
0
−0.2
−0.4
−0.6
10
5
10
5
0
0
y
−5
−5
x
−10
−10
Fig. 12.20
Approximation of the test function of Eq. (
12.27
), using a one hidden layer FNN with
sigmoidal basis functions
networks and OHL-FNN succeed better approximation than RBF. In the case of
the 2D functions, the performance of OHL-FNN improved when more nodes were
added to the hidden layer, while the performance of RBF improved when the spread
of the gaussians was increased.
