148
a)
X(t-1)
X(t-2)
X(t-q)
b)
Input
layer
0
0
0
0
0
0
• • •
0
0
0
Input
layer
T.-S. Chon . Y.S. Park· I.-S. Kwak . E.Y. Cha
~i
Hidden
layer
0
0
Wkj
0
0
0
Hidden
layer
Output
layer
0
0
0
Target [X(~]
Error
Output
layer
Output
[Y(~]
XI(~' ... , XN(~
_--:;;:;1If.J'~--_..r )---'-""'" YI( ~
~=--':>'~'--!;.~~~-:---/-~~--1....,.. Y2( ~
M
Context
layer
~(t-1)
Fig. 8.12. The architecture of artificial neural networks with temporal patterning.
(Prom Chan et al. 2000b). a) time delayed multilayer perceptron. b) Elman type
recurrent neural network (RNN).