Consider a dataset where two classes are linearly separable. The decision boundary is given as w1 Ty + wexs + b= 0withw; =2, ws = 1,and b = -3. 1. Compute the margin for the hyperplane if the support vectors are (3, 1) and (2, 5). 2. Determine whether the point (4, 4) is correctly classified. If not, compute the hinge loss for this point. A neural network has a single hidden layer with two neurons, each using a ReLU activation function. The weights of the network are as follows: -1.0 2.0 0.5 Hidden layer to output: W = [1.0 0.5], bias b = 0.3 e Input to hidden layer: W = [ & 1'0], bias b = [0'2] For an input X = [g] 1. Compute the output of the hidden layer after applying RelLU activation. 2. Compute the final output of the network. Given a dataset with one feature X = [1,2, 3, 4] and target Y = (2, 3, 5, 7], you are training a linear regression model y = By + 31 z. 1. Initialize By = 0 and 3; = 0. Using a learning rate of 0.1, compute the first two steps of gradient descent manually. 2. Write the cost function after each step