$($a$)$ Consider a two input XNOR gate and simulate a perceptron algorithm for it$,$ where,

learning rate $=0\mathrm{.}02$ and threshold $=0\mathrm{.}2.$

$($b$)$ Represent using a multilayer neural network $Ao.Bo.Co.Do.E$ where $o.$ represents

XNOR. What will be the optimal depth and width for this network.

$($c$)$ You'd like to train a fully$-$connected neural network with $7$ hidden layers, each with

$8$ hidden units. The input is $30-$dimensional and the output is a binary. What is the

total number of trainable parameters in your network?