Question

For the below neural network, use your roll number and calculate the initial values of i1, i2, w1, w2, w3, w4, w5, w6, w7, w8, b1, b2. As an example, the below table shows calculations for a sample roll number ‘123456’.
w1 w2 w3 w4 i1 i2
(1+1)/10
= 0.2
(1+2)/10
= 0.3
(1+3)/10
= 0.4
(1+4)/10
= 0.5
(1+5)/10
= 0.6
(1+6)/10
= 0.7
w5 w6 w7 w8 b1 b2
(6+1)/10
= 0.7
(6+2)/10
= 0.8
(6+3)/10
= 0.9
(6+4)/10
= 1.0
(6+5)/10
= 1.1
(6+6)/10
= 1.2
Further to calculating the initial values of the weights, with values o1 = 0.01; o2 = 0.99; α = 0.5 (learning rate); and activation function a sigmoid function, do the following
a. Use forward propagation and calculate the total error. .{1 mark}
b. Use backward propagation and calculate updated values of w5, w6, w7, w8.{2 marks}
c. Use backward propagation and calculate updated values of w4, w3, w2, w1

Answer 1

Answer:

put the questions Clearly