MCQ

Question 1

Which statement is true about 1-vs-1 classification?

Answer 1

It needs as many models as there are class pairs, and each model predicts which class an observation is more likely to belong to

Question 2

Which statement is true about 1-vs-all classification?

Answer 2

It needs as many models as there are classes, and each models predicts the probability of an observation to belong to a class

Question 3

In a binary classification setting, accuracy is…

Answer 3

The ratio of number of correct predictions to the number of observations

Question 4

An advantage of using the bootstrap method over other model selection methods is….

Answer 4

It uses fewer data points for training and so each model is more accurate

Question 5

The classical estimate of the model error obtained using the bootstrap method is

Answer 5

It overestimates the real error, because each model is trained on average on 63.2% of the data set

Question 6

What are typical assumptions of the error term ε, when the relation between inputs and
output is written as y = f(X) + ε (where y is the output, X is the input, and f is the
real relationship between input and output)?

Answer 6

It’s expected value is zero

Question 7

Which one of the following about the F1 score is true?

Answer 7

It is close to zero for models bad at discriminating the positive class

Question 8

Given data with 3 numeric features, 2 categorical features and 1 numeric label, how many subset of features should I consider, if I do feature selection via an exhaustive search of all possible combinations?

Answer 8

32

Question 9

What is the synonym of “feature”?

Answer 9

Independent variable

Question 10

What is true about forward stepwise feature selection?

Answer 10

It starts with considering no features, just predicting mean label

Question 11

The gradient descent method

Answer 11

Moves against the direction of gradient in every iteration

Question 12

In the hold out validation method

Answer 12

We divide our data into large training set and small test set

Question 13

After selecting a winning model in the hold out validation method, it’s a good idea to

Answer 13

Retraining the training+test data before using it in production

Question 14

The estimate of the error of the model we get using the hold out validation method

Answer 14

Is an overestimation of the true model error because we train on fewer data points

Question 15

The hyperparameters of a model

Answer 15

Are chosen by the user usually by hyper parameter tuning

Question 16

A hyperplane in R
p
can be described with…

Answer 16

One linear equation

Question 17

What is true about irreducible error?

Answer 17

It depends on the data noise and cannot be reduced by selecting a better model

Question 18

The k-fold cross validation when k=n (n is the number of data points), becomes

Answer 18

The leave-one-out cross-validation method

Question 19

An advantage of the k-fold cross validation is that

Answer 19

it allows a trade off between accuracy and computational power required which we can tune by varying k

Question 20

Increasing k in k-fold cross-validation is going to

Answer 20

reduce the bias of the estimate of the model error, because training sets get larger

Question 21

Which of the following properties are commonly ascribed to Lasso Regularisation?
(a) It increases the variance of a model but, in exchange, drastically reduces the bias
(b) It penalises simple models, with few parameters
(c) It can help the interpretability of the model, because it tends to set many parameters
to zero
(d) It makes training a model easier by smoothing out irregularities in the objective
function

Answer 21

It can help improve the interpretability as it sets many parameters to zero

Question 22

Which of the following statement about Lasso regularisation is true?

(a) It penalises the number of non-zero parameters
(b) It penalises the 1-norm of the parameter vector (i.e., the sum of absolute values)
(c) It penalises the 2-norm of the parameter vector (i.e., the sum of squares)
(d) It penalises the infinite-norm of the parameter vector

Answer 22

It penalises the 1-norm of the parameter vector

Question 23

Which of the following about Lasso and Ridge Regularisation is true?
(a) Lasso always dominates Ridge regularisation, giving smaller test errors
(b) Ridge always dominates Lasso regularisation, giving smaller test errors
(c) In general neither method dominates the other, although there are rules of thumb
on when we might expect dominance
(d) Neither method dominates the other in terms of test error, but Lasso always gives
smaller training errors

Answer 23

In general neither method dominates the other although there are thumb rules

Question 24

As a rule Lasso tends to perform better than Ridge when

(a) A small number of predictors have large coefficients, and the remaining ones have coefficients that are very small
(b) The response is a function of many predictors, all with coefficients of roughly equal size
(c) A small number of predictors have negative coefficients, and the remaining ones have coefficients that are very small and positive
(d) A small number of predictors have positive coefficients, and the remaining ones have coefficients that are very small and negative

Answer 24

A small no. of predictors have large coefficients and the remaining ones have coefficients that are very small

Question 25

In gradient descent, the learning rate

(a) Indicates by how much we move against the gradient
(b) Helps forgetting about outliers and concentrating on typical samples
(c) Can be chosen appropriately, so to normalise the inputs
(d) Has no impact on the number of iterations required for convergence

Answer 25

Indicates by how much we move against the gradient

Question 26

Two classes are called linearly separable if

(a) There is a third class which lies between them
(b) They are separated by a marin wider than the average inter-class distance
(c) They both lie on the same line
(d) There is a hyperplane that separates them

Answer 26

There is a hyperplane that separates them

Question 27

At a local minimum of the average loss function

(a) The gradient with respect to the model’s parameters is zero
(b) The gradient with respect to the model parameters is strictly positive
(c) The gradient with respect to the model parameters is strictly negative
(d) The gradient with respect to the model’s parameters can have any sign

Answer 27

The gradient with respect to the model’s parameters is zero

Question 28

The estimate of the accuracy of a model obtained using Leave-One-Out Cross-Validation

(a) Higher bias than the estimate provided by hold-out cross-validation because we test on a single point
(b) Higher bias than the estimate provided by k-fold cross-validation because we test on a single point
(c) Higher bias than the estimate provided by hold-out cross validation because we train on a smaller set
(d) Lower bias than the estimate provided by hold-out cross validation because we train on a larger set

Answer 28

Lower bias than the estimate provided by hold-out cross validation because we train on a larger set

Question 29

A disadvantage of the Leave-One-Out Cross-Validation is that

(a) The estimate of the model’s error we obtain is a gross overestimation of the real error
(b) The estimate of the model’s error we obtain is a gross underestimation of the real error
(c) The estimate of the model’s error we obtain is affected by roughly twice as much variance compared to k-fold CV
(d) We have to train each model as many times as we have points in the dataset, and this can be computationally expensive

Answer 29

We have to train each model as many times as there are points in the dataset and this can be computationally expensive

Question 30

Which of the following is a desirable property of the loss function?

(a) It achieves its minimum when the prediction is correct
(b) It has many discontinuities, to facilitate finding a global optimum
(c) It’s neither concave nor convex
(d) It decreases, the further the prediction is compared to the ground truths

Answer 30

It achieves its minimum when the prediction is correct