An axis$-$aligned n$-$dimesional rectangle classifier h$($l$,$u$)$ is given by two vectors l$,$ u in R n such that l $<$ u $($i$.$e$.$ li $<$ ui for all $1$ i $$ n$).$ A vector x in R n is labeled $1$ by this classifier h$($l$,$u$)$ if l $<$ x $<$ u$,$ i$.$e$.$ for every i holds li $<$ xi $<$ ui $.$ Otherwise x is labeled $0.$ Problem B $[2$ pts$]$ You are tasked with prediction of a heart attack inpatients. A training set consists of data points $($patient$)$ that have the following features: $$ blood pressure $($BP$),$ body$-$mass index $($BMI$),$ age $($A$),$ level of physical activity $($P$),$ income $($I$).$ For legacy reasons you have to pick between $2$ algorithms: the first algorithm ERM picks an axis aligned rectangle in the two dimensional space spanned by the features BP and BMI and the other algorithm ERM picks an axis aligned rectangle in the five$-$dimensional space spanned by all features. $1$pt Explain the pros and cons of each choice $1$pt Explain how the number of available labeled training samples will affect your choice Hint: Think of SRM paradigm when you are trying to give your reasons