The mathematical foundations of machine learning are linear algebra, calculus, probability and statistics, the tools used to describe and optimize models. Questions define scalars, vectors, matrices and tensors, and test dot products, matrix multiplication, transposes, eigenvalues and eigenvectors. Calculus items cover derivatives, partial derivatives that hold other variables constant, the chain rule and gradients used in gradient descent. Probability questions address distributions such as normal, Bernoulli and binomial, expectation, variance, conditional probability and Bayes' rule. Advanced items ask about entropy, cross-entropy, Kullback-Leibler divergence between distributions, maximum likelihood estimation and the Fisher information matrix.
Below are 30 practice questions from a pool of 210 Mathematical Foundations MCQs, one of 17 topics in AI / Machine Learning & Data Analytics. Each shows the correct answer with an explanation; when you are ready, take a timed quiz to test recall under exam conditions.
Each question below shows the correct answer with a full explanation. Use these to build conceptual understanding before attempting a timed quiz.
Q1. What is a scalar in linear algebra?
A scalar is a single numerical value, as opposed to vectors, matrices, or higher-order tensors.
Q2. What is the mean of the numbers 2, 4, 6, 8, 10?
The mean is (2+4+6+8+10)/5 = 30/5 = 6.
Q3. A matrix with equal rows and columns is called:
A square matrix has the same number of rows and columns.
Q4. What does probability measure?
Probability measures the likelihood or chance of an event occurring, ranging from 0 to 1.
Q5. The derivative of a constant is:
The derivative of a constant is always 0 because a constant has no rate of change.
Q6. What is the dot product of vectors [1,2] and [3,4]?
Dot product = (1*3) + (2*4) = 3 + 8 = 11.
Q7. What is the median of {1, 3, 5, 7, 9}?
The median is the middle value in a sorted set. In {1,3,5,7,9}, the middle value is 5.
Q8. An identity matrix has:
An identity matrix has 1s on the main diagonal and 0s in all other positions.
Q9. The range of a probability value is:
Probability values always range from 0 (impossible) to 1 (certain).
Q10. What is the transpose of a matrix?
The transpose of a matrix is obtained by interchanging its rows and columns.
Q11. The gradient of a function points in the direction of:
The gradient vector points in the direction of the steepest increase of the function.
Q12. Bayes' theorem relates:
Bayes' theorem describes how to update the probability of a hypothesis given new evidence, relating prior and posterior probabilities.
Q13. The eigenvalue equation is Av = λv. What is λ?
In the equation Av = λv, λ is the eigenvalue, a scalar that indicates how the eigenvector v is scaled by matrix A.
Q14. What is the chain rule used for in calculus?
The chain rule is used to compute the derivative of a composite function: d/dx[f(g(x))] = f'(g(x)) * g'(x).
Q15. Standard deviation measures:
Standard deviation measures how spread out data values are from the mean of the dataset.
Q16. A positive definite matrix has:
A positive definite matrix has all positive eigenvalues, which is important in optimization (guarantees a minimum).
Q17. In probability, two events are independent if:
Two events are independent if the probability of both occurring equals the product of their individual probabilities.
Q18. The determinant of a 2x2 matrix [[a,b],[c,d]] is:
The determinant of a 2x2 matrix is calculated as ad - bc.
Q19. A convex function has the property that:
A key property of convex functions is that any local minimum is also a global minimum, which simplifies optimization.
Q20. The covariance between two identical variables equals:
Cov(X,X) = Var(X). The covariance of a variable with itself is its variance.
Q21. The Hessian matrix contains:
The Hessian matrix contains all second-order partial derivatives of a scalar-valued function, used to analyze curvature.
Q22. The Kullback-Leibler divergence measures:
KL divergence measures the difference between two probability distributions, commonly used in ML for comparing distributions.
Q23. A Jacobian matrix represents:
The Jacobian matrix contains all first-order partial derivatives of a vector-valued function with respect to another vector.
Q24. In the context of ML, why is the log-likelihood used instead of likelihood?
Log-likelihood converts the product of probabilities into a sum, making numerical computation more stable and differentiation easier.
Q25. The singular value decomposition (SVD) decomposes a matrix into:
SVD decomposes a matrix A into UΣV^T where U and V are orthogonal matrices and Σ is a diagonal matrix of singular values.
Q26. A saddle point in optimization is where:
At a saddle point, the gradient is zero but the point is a minimum along some directions and a maximum along others.
Q27. The moment generating function uniquely determines:
If it exists in a neighborhood of zero, the moment generating function uniquely determines the probability distribution.
Q28. In matrix calculus, the gradient of x^T A x with respect to x is:
The gradient of x^T A x with respect to x is (A + A^T)x. If A is symmetric, this simplifies to 2Ax.
Q29. The condition number of a matrix indicates:
The condition number measures how sensitive a matrix computation is to perturbations in the input, important for numerical stability.
Q30. Jensen's inequality states that for a convex function f:
Jensen's inequality states that for a convex function, the function of the expected value is less than or equal to the expected value of the function.
Practice MCQs with explanations
Practice MCQs with explanations
Practice MCQs with explanations
Practice MCQs with explanations
Practice MCQs with explanations
Practice MCQs with explanations
Take a timed quiz drawn from 210+ questions on this topic. No signup required — your progress saves in your browser.
Start Mathematical Foundations Quiz