Machine Learning

Question 1

Method for computer-aided, automated learning of models that
represent relationships between the features 𝑋1, … , 𝑋𝑝 and the
associated target variable 𝑌 .

Answer 1

Machine Learning

Question 2

The three types of learning that Machine Learning combines.

Answer 2

Unsupervised learning, supervised learning and reinforcement learning.

Question 3

The two types of supervised learning tasks.

Answer 3

Classification and regression

Question 4

A qualitative / categorical supervised learning task is a ___ task.

Answer 4

Classification

Question 5

Which type of task should we use to predict the outcome of a COVID test (positive or negative), a spam filter, the credit worthiness of a customer, the result of a match (win, tie, lose)?

Answer 5

A Classification task

Question 6

A quantitative / numerical supervised learning task is a ___ task.

Answer 6

Regression

Question 7

Which type of task should we use to predict mileage of electric cars, traffic volume at a street segment, expected waiting time (e.g., until e-scooter is fully charged), climate change (incl. global warming, or emission of greenhouse gases), amount of energy that is produced by wind parks?

Answer 7

A Regression task

Question 8

The two machine learning intentions.

Answer 8

Prediction and explanation

Question 9

It defines what should be modeled within the ML pipeline, it comprises feature data and target variable.

Answer 9

A task

Question 10

An algorithm that trains the model based on training data, it encapsulates the actual supervised learning model.

Answer 10

A learner

Question 11

It is an instantiation of the learner to the data (with model parameters that are optimal for the given data).

Answer 11

A model

Question 12

(True or false) Hyperparameters are defined within the learner.

Answer 12

True. A learner is a (data independent) general construct defining which ML algorithm (and
hyperparameters) will be used.

Question 13

Performance measure that quantifies deviation between true and predicted values of the target variable.

Answer 13

Loss function

Question 14

The aggregation of all losses across (training) data.

Answer 14

Empirical risk

Question 15

The best parametrization of the learner when training a model.

Answer 15

Risk minimization

Question 16

MLR3 learner in which for each (test) data point x = (𝑥1, … , 𝑥𝑝), i.e., each observation for which we want to make a prediction, compute the neighborhood 𝑁𝑘 (𝑥1, … , 𝑥𝑝) of its 𝑘 nearest
neighbors.

Answer 16

K Nearest Neighbor Classifier

Question 17

(True or false) There is no training step for k-NN models, just storing the training data to process it during the predict step.

Answer 17

True

Question 18

If a model has learned too many details from the training data, which don’t necessarily generalize to unseen data sets, it is called an ____ model.

Answer 18

Overfitted

Question 19

If a model is too general to capture all the relevant patterns from the data, then it is called an ____ model.

Answer 19

Underfitted

Question 20

What kind of fit does a model have if it has low error in the training data as well as in the test data.

Answer 20

Underfit

Question 21

One of the main disadvantages of this learner is that it can be computationally expensive for large datasets as it needs to store a lot of data per observation.

Answer 21

KNN Learner

Question 22

Trick that supports classifiers as knn and SVMs to map the problem into a higher-dimensional space, then classifies the transformed problem and then use the outcome of the aforementioned classification for the original problem.

Answer 22

The kernel trick

Question 23

The name given to data in which classes are distributed unevenly.

Answer 23

Imbalanced data.

Question 24

Two typical countermeasures to handling unbalanced data.

Answer 24

• Over-/undersampling,
• Tuning class thresholds

Question 25

A performance measurement for machine learning classification problem where output can be two or more classes. It is a table with 4 different combinations of predicted and actual values.

Answer 25

Confusion matrix

Question 26

A plot of test sensitivity as the y coordinate versus its specificity or false positive rate (FPR) as the x coordinate, is an effective method of evaluating the performance of diagnostic tests.

Answer 26

Receiver Operating Characteristics (ROC) Curve

Question 27

In the receiver operating characteristics (ROC) curve, the ideal classifier lies at the ___ corner.

Answer 27

Top left

Question 28

(True or false) In the context of regression larger values of R2 adjusted are better.

Answer 28

True

Question 29

(True or false) Decision boundaries are defined for regression problems.

Answer 29

False. Decision boundaries are defined for classification problems only.

Question 30

(True or false) Both supervised and unsupervised machine learning models can have hyperparameters.

Answer 30

True

Question 31

When your data is noisy, what would you do if you wanted to apply k-NN?

Answer 31

Increase k