uiz Space

May 2026 term · Machine Learning Techniques · BSCS2007

Machine Learning Techniques End Term: 13 September 2026, Set S2 (May 2026 term)

The IIT Madras BS Machine Learning Techniques (MLT) End Term paper sat on 13 Sept 2026, in the May 2026 term, set S2: 18 questions for 47 marks in 180 minutes. Every question is below with its answer. Take it as a timed mock test to be marked, or read it through first.

Questions
18
Marks
47
Duration
180 min
MSQ
7
Numerical
4
MCQ
7
Take as mock testPractise with answers

Updated

Official paper: Machine Learning Techniques 13 Sep 26 (Session 2) · No negative marking.

Question 1

+4 marksOne or more correct options

Select all that apply.

  1. A
  2. B
  3. C
  4. D
Show answer

Correct answers

  • C
  • D
Question 1 on its own page

Question 2

+4 marksOne or more correct options

Select all that apply.

  1. A

    Perceptron.

  2. B
  3. C
  4. D

    A decision tree using axis-aligned splits.

  5. E

    Gaussian Naïve Bayes in which the class-conditional feature variances differ between the two classes.

  6. F

    A support vector machine with a linear kernel.

  7. G

    A support vector machine with an RBF (Gaussian) kernel.

Show answer

Correct answers

  • C
  • D

    A decision tree using axis-aligned splits.

  • E

    Gaussian Naïve Bayes in which the class-conditional feature variances differ between the two classes.

  • G

    A support vector machine with an RBF (Gaussian) kernel.

Question 2 on its own page

Question 3

+2 marksOne or more correct options

Let

Select all that apply.

  1. A
  2. B
  3. C
  4. D
Show answer

Correct answers

  • A
  • C
Question 3 on its own page

Question 4

+2 marksOne or more correct options

Select all that apply.

  1. A
  2. B
  3. C
  4. D
Show answer

Correct answers

  • A
  • D
Question 4 on its own page

Question 5

+2 marksOne or more correct options

Consider the effect of each operation below on a model's bias and variance. Which of the following statements are correct?

Select all that apply.

  1. A

    Regularising the weights in a linear/logistic regression model increases bias and decreases variance.

  2. B

    Increasing the maximum depth of a decision tree increases bias and decreases variance.

  3. C

    Increasing the number of hidden units in a neural network decreases bias and increases variance.

  4. D

    Removing all the non-support-vectors from the training set of an SVM leaves both bias and variance unchanged.

Show answer

Correct answers

  • A

    Regularising the weights in a linear/logistic regression model increases bias and decreases variance.

  • C

    Increasing the number of hidden units in a neural network decreases bias and increases variance.

  • D

    Removing all the non-support-vectors from the training set of an SVM leaves both bias and variance unchanged.

Question 5 on its own page

Question 6

+2 marksNumerical answer
Show answer

Correct answer: 1.09 (accepted within ±0.01)

Question 6 on its own page

Question 7

+2 marksNumerical answer
Show answer

Correct answer: 180

Question 7 on its own page

Question 8

+4 marksOne correct option
  1. A
  2. B
  3. C
  4. D
Show answer

Correct answer

  • A
Question 8 on its own page

Question 9

+4 marksOne correct option
  1. A

    (A figure from the original paper is missing from the source site.)

  2. B

    (A figure from the original paper is missing from the source site.)

  3. C

    (A figure from the original paper is missing from the source site.)

  4. D
Show answer

Correct answer

  • A

    (A figure from the original paper is missing from the source site.)

Question 9 on its own page

Question 10

+3 marksNumerical answer

Consider one-dimensional ridge regression with no bias term. Given

and

Show answer

Correct answer: 0.2 (accepted within ±0.01)

Question 10 on its own page

Question 11

+1 markOne correct option

For the covariance matrix

, compute the variance captured along the first principal component.

  1. A
  2. B
  3. C
  4. D
Show answer

Correct answer

  • C
Question 11 on its own page

Question 12

+1 markOne correct option
  1. A
  2. B
  3. C
  4. D
Show answer

Correct answer

  • B
Question 12 on its own page

Question 13

+2 marksOne correct option
  1. A
  2. B
  3. C
  4. D
Show answer

Correct answer

  • C
Question 13 on its own page

Question 14

+2 marksOne correct option

Consider a binary classification dataset in which the two classes are perfectly linearly separable. You are considering a Perceptron, Logistic Regression, and a Decision Tree. If the objective is to learn a linear decision boundary and no probability estimates are required, which of the following classifiers are appropriate choices?

  1. A

    Perceptron only

  2. B

    Logistic Regression only

  3. C

    Both Perceptron and Logistic Regression

  4. D

    Decision Tree only

Show answer

Correct answer

  • C

    Both Perceptron and Logistic Regression

Question 14 on its own page

Question 15

+2 marksOne correct option
  1. A
  2. B
  3. C
  4. D
Show answer

Correct answer

  • C
Question 15 on its own page

Question 16

+4 marksNumerical answer

Note: coordinates of x1 vector are intentionally not provided.

Show answer

Correct answer: 2.72 (accepted within ±0.01)

Question 16 on its own page

Question 17

+3 marksOne or more correct options

Which of the following statements about entropy and information gain in decision-tree learning are correct?

Select all that apply.

  1. A

    If a feature splits a node into perfectly pure children, its information gain at that node equals the entropy of that node.

  2. B
  3. C

    Information gain can be negative if a split increases impurity.

  4. D

    After a parent node is split into child nodes, no individual child node can have entropy greater than the parent node's entropy.

Show answer

Correct answers

  • A

    If a feature splits a node into perfectly pure children, its information gain at that node equals the entropy of that node.

  • B
Question 17 on its own page

Question 18

+3 marksOne or more correct options

Which of the following statements about bagging are correct?

Select all that apply.

  1. A

    Bagging reduces the variance of the ensemble's prediction.

  2. B

    Bagging substantially reduces the bias of the base learner.

  3. C

    Bagging tends to help most when the base learners are diverse (weakly correlated with one another).

  4. D

    Bagging tends to help most with high-variance, low-bias base learners such as deep, unpruned decision trees.

  5. E

    averaging all base learner reduces variance, if every base learner produces identical predictions.

Show answer

Correct answers

  • A

    Bagging reduces the variance of the ensemble's prediction.

  • C

    Bagging tends to help most when the base learners are diverse (weakly correlated with one another).

  • D

    Bagging tends to help most with high-variance, low-bias base learners such as deep, unpruned decision trees.

Question 18 on its own page