Question No $01$:

Consider the neural network architecture for XAND function given below:

Assume that the activation function is the sigmoid function. Initialize all

the weights with $0\mathrm{.}1.$ Write down the values of the weights after the first

and the second iteration of Backpropagation algorithm run with the

following examples:

$`$onsider a neural network that uses a function tanh instead of the sigmoid

unction. What will be the values of the weights after the first and second

eration in such a case?