Suppose We have ANN with 3 layers. Input layer with 2 input neurons. Hidden layer with 2 neurons and output layer with 1 neuron. Hidden and output neurons are using sigmoid function as an activation function

Initial weights are as following:

w1 = 0.11, w2 = 0.21, w3 = 0.12, w4 = 0.08, w5 = 0.14 and w6 = 0.15. (Note: Bias is zero for each neuron)

Single sample is as following:

inputs=[2, 3] and output=[1].

Perform the following task:

1. 1. Forward propagation to identify predicted value.
2. 2. Calculate prediction error
3. 3. Perform back propagation to identify updated values for w1, w2, w3, w4, w5, and w6. For back propagation learning rate=0.05.

Input layer Hidden Layer Output layer wi hi prediction W2 Ws out W6 W3 h WA