Binary Classification With Surrogate Loss Function Consider the problem of binary classification based on the following simple training data: (z3, ) = (4, +1). Assume we will construct a classifier of form y - sign(x -b), where b is a parameter to be estimated, (a) 5 pointsl Define the 0/1 loss Lo/1(b) to be the number of mistakes the classifier with param- eter b makes on the training data. Specifically, where (t)-1 if t 0 and 0 if t > 0. Please plot its curve (with x-axis being b and y-axis the value of Lo/1(b)). Set the range of b to be [0,5]. What is the (set) of optimal parameters b according to this loss'? Binary Classification With Surrogate Loss Function Consider the problem of binary classification based on the following simple training data: (z3, ) = (4, +1). Assume we will construct a classifier of form y - sign(x -b), where b is a parameter to be estimated, (a) 5 pointsl Define the 0/1 loss Lo/1(b) to be the number of mistakes the classifier with param- eter b makes on the training data. Specifically, where (t)-1 if t 0 and 0 if t > 0. Please plot its curve (with x-axis being b and y-axis the value of Lo/1(b)). Set the range of b to be [0,5]. What is the (set) of optimal parameters b according to this loss