A neural network is designed to approximate a function $f(x)=sin(x)$ using a multilayer

perceptron $($MLP$)$ with a single hidden layer. The neural network has one input neuron,

$n$ neurons in the hidden layer, and one output neuron.

$[4$ Marks$]$

Explain the process of training this MLP using backpropagation to minimize the

error between the predicted output and the actual function values. include the

steps involved in forward propagation.

Discuss the impact of varying the number of neurons $(n)$ in the hidden layer on

the MLP$\text{'}$s ability to approximate the sine function accurately. Provide insights

into the trade$-$offs associated with increasing or decreasing the number of

neurons in the hidden layer.