Wayground logo

Free Printable Worksheets

Font size

S
M
L
XL
Worksheets

Machine Learning and Linear Algebra Quiz

Total questions: 20

Worksheet time: 10mins

Name
Class
Date
1.

Which of the following best describes the experience E in supervised learning?

a)

A model trained without any label

b)

A dataset containing only input features

c)

A dataset of (input, label) pairs used for learning

d)

The performance metric used to evaluate models

2.

In the context of ML model evaluation, which metric is not derived from a confusion matrix?

a)

F1 Score

b)

Precision

c)

Recall

d)

Log-likelihood

3.

Empirical risk is minimized during training. What does it represent?

a)

The model's confidence in predictions

b)

The average loss on the training data

c)

The regularization term added to prevent overfitting

d)

The variance of the prediction errors

4.

Which of the following is not an axiom of probability?

a)

Non-negativity

b)

Independence

c)

Total probability equals 1

d)

Additivity

5.

Bayes' theorem can be used to:

a)

Estimate model weights

b)

Convert prior probabilities to posterior probabilities

c)

Normalize feature vectors

d)

Find eigenvectors of the covariance matrix

6.

The matrix A is invertible if:

a)

Its rank is less than n

b)

Its determinant is zero

c)

Its null space is non-trivial

d)

There exists a matrix A⁻¹ such that AA⁻¹=I

7.

Which of the following is true about the dot product u⋅v?

a)

It's the sum of the element-wise division

b)

It equals the L2 norm of the vector

c)

It is zero if u and v are orthogonal

d)

It can only be computed for square matrices

8.

What does the rank of a matrix correspond to?

a)

The number of zero rows

b)

The dimension of its column space

c)

The number of elements in the null space

d)

The number of pivots in its transpose

9.

The condition number of a matrix gives insight into:

a)

The number of non-zero eigenvalues

b)

How sparse the matrix is

c)

The matrix’s numerical stability when solving Ax=b

d)

The number of dimensions in its eigenspace

10.

If a matrix A is symmetric and positive definite, then:

a)

All its eigenvalues are negative

b)

It has complex eigenvalues

c)

It is guaranteed to be invertible

d)

Its rank is always 1

11.

Which supervised learning algorithm uses information gain to build its structure?

a)

Logistic Regression

b)

K-Nearest Neighbors

c)

Decision Tree

d)

Support Vector Machine

12.

What does the entropy of a node in a decision tree measure?

a)

The number of features selected

b)

The variance in feature values

c)

The uncertainty or impurity in the class distribution

d)

The maximum likelihood of prediction

13.

In K-Nearest Neighbors, increasing the value of k usually leads to:

a)

Higher variance

b)

Lower bias

c)

Better performance on all datasets

d)

Smoother decision boundaries

14.

Which of the following statements about logistic regression is true?

a)

It minimizes squared error

b)

It produces multi-class outputs by default

c)

It models the probability using a sigmoid function

d)

It uses k-nearest neighbors for estimation

15.

Support Vector Machines aim to:

a)

Minimize classification error directly

b)

Maximize the distance between support vectors

c)

Maximize the geometric margin

d)

Minimize the entropy of the decision boundary

16.

In K-means clustering, the objective is to minimize:

a)

The silhouette score

b)

The variance between clusters

c)

The within-cluster sum of squared distances

d)

The information gain

17.

What does the elbow method help determine?

a)

The cluster initialization points

b)

The number of relevant features

c)

The optimal number of clusters K

d)

The performance of agglomerative clustering

18.

In PCA, the principal components are:

a)

Randomly chosen from the dataset

b)

Orthogonal vectors capturing maximum variance

c)

Cluster centers from K-means

d)

Linear regression coefficients

19.

Which of the following is not a typical use case for PCA?

a)

Dimensionality reduction

b)

Feature decorrelation

c)

Supervised classification

d)

Data visualization

20.

The eigenvectors of the covariance matrix in PCA:

a)

Define the directions of least variance

b)

Form a dependent basis for the data

c)

Are ranked by the size of their corresponding eigenvalues

d)

Are always complex-valued