Page 41 - Algorithm Collections for Digital Signal Processing Applications using MATLAB

1. Artificial Intelligence 29

w 31 (n+1) = w 31 (n) + Δw 31(n+1) + ρ H * Δw 31(n)

b h (n+1)= b h (n) +e* γ H + ρ H * Δb h (n)

Δ b h (n+1)

b 1 (n+1) = b 1 (n) + Δ b 1 (n+1) + ρ O * Δb 1 (n)

b 2 (n+1) = b 2 (n) + Δ b 2 (n+1) + ρ O * Δb 2 (n)

ρ H and ρ O are the momentums used in the hidden and output layer
respectively.

4.3 Example

Consider the problem for training the Back propagation Neural Network
with Hetero associative data as mentioned below

Input Desired Output

0 0 0 0 0
0 0 1 0 1
0 1 0 0 1
0 1 1 0 1
1 0 0 0 1
1 0 1 0 1
1 1 0 0 1
1 1 1 1 1

ANN Specifications

Number of layers = 3
Number of neurons in the input layer = 3
Number of neurons in the hidden layer =1
Number of neurons in the output layer = 2
Learning rate =0.01
Transfer function used in the hidden layer = ‘logsig’
Transfer function used in the output layer = ‘linear’ (i.e) output is taken as
obtained without applying non-linear function like ‘logsig’,’ tansig’.