q$2)$

Answer the below with true$/$false and justify in at most $2$ sentences $[6$ Marks$]$

A$.$ True$/$False: Using a model with less bias is always better than using a model with more bias. Explain.

B$.$ True$/$False: With the correct step size, gradient descent always converges to the optimum of the objective function for linear regression if the optimum exists. Explain.

C$.$ True$/$False: Logistic regression can only be used for binary classification problems.

D$.$ True$/$False: Gradient descent can only be used for convex functions.

E$.$ True$/$False: In classification problems, the Cross$-$Entropy Loss is a suitable cost function.

F$.$ True$/$False To predict the probability of an event, one would prefer a regression model trained with squared error to a classifier trained with logistic regression