Question 24: A two-layer neural network with the D features, x_1,\ x_…

A two-layer neural network with the $D$ features, $x_1,\ x_2,\ \ldots,\ x_D$ is defined below:

$\begin{aligned} h_1 &= W_1 x + b_1 \\ z_1 &= ReLU(h_1) \\ h_2 &= W_2 z_1 + b_2 \\ \hat{y} &= \sigma(h_2) \\ \ell &= -y.log(\hat{y}) - (1 - y).log(1 - \hat{y}) \end{aligned}$

compute $\frac{\partial h_2}{\partial b_2}$, $\frac{\partial z_1}{\partial h_1}$ and $\frac{\partial h_1}{\partial W_1}$ .