Figure from the original question paper

(C) Option figure from the original question paper

(A) Option figure from the original question paper

(B) Option figure from the original question paper

(D) Option figure from the original question paper

Question text with a plot of + and - training points, a decision boundary and weight vector w for a hard-margin SVM question

(B) w shown in this diagram is not the optimal weight vector for a hard-margin, linear-SVM

(A) w shown in this diagram is the optimal weight vector for a hard-margin, linear- SVM

Figure from the original question paper

(B) Option figure from the original question paper

(D) Option figure from the original question paper

(A) Option figure from the original question paper

(C) Option figure from the original question paper

Figure from the original question paper

(A) Option figure from the original question paper

(D) Option figure from the original question paper

(B) Option figure from the original question paper

(C) Option figure from the original question paper

Question text with a plot of two loss curves L(t) against t labelled (1) and (2) for training and validation loss

(A) (1) is the loss on the validation dataset

(B) (2) is the loss on the training dataset

(C) (1) is the loss on the training dataset

(D) (2) is the loss on the validation dataset

Figure from the original question paper

(B) Option figure from the original question paper

(C) Option figure from the original question paper

(A) Option figure from the original question paper

(D) Option figure from the original question paper

Figure from the original question paper

(B) Option figure from the original question paper

(C) Option figure from the original question paper

(A) Option figure from the original question paper

(D) Option figure from the original question paper

January 2023 term · Machine Learning Techniques · BSCS2007

MLT End Term: 30 April 2023, Set QPD1-S1 (January 2023 term)

The IIT Madras BS Machine Learning Techniques (MLT) End Term paper sat on 30 Apr 2023, in the January 2023 term, set QPD1-S1: 23 questions for 100 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: 23
Marks: 100
Duration: 180 min
MCQ: 7
MSQ: 7
Numerical: 9

Take as mock test Practise with answers

Updated 25 Sept 2026

Official paper: IIT M DIPLOMA ET1 EXAM QPD1 S2 30 Apr 2023 · No negative marking.

Question 1

+5 marksOne correct option

Show answer

Correct answer

Question 1 on its own page

Question 2

+5 marksOne correct option

Show answer

Correct answer

Question 2 on its own page

Question 3

+5 marksOne correct option

A
Overfitting
B
Underfitting

Show answer

Correct answer

A
Overfitting

Question 3 on its own page

Question 4

+5 marksOne correct option

Consider the following training dataset for a binary classification problem in $\mathbb{R}^2$ :

$x_1$	$x_2$	$y$
$1$	$2$	$1$
$1$	$-2$	$1$
$-5$	$0$	$1$
$5$	$0$	$-1$
$-2$	$1$	$-1$
$-2$	$-1$	$-1$

If we try to learn a perceptron model for this dataset, will the algorithm ever converge to a weight vector? Select the most appropriate answer with the information available to you.

A
Yes, it will certainly converge to a weight vector.
B
No, it will never converge.

Show answer

Correct answer

B
No, it will never converge.

Question 4 on its own page

Question 5

+5 marksOne correct option

While training a perceptron model, the weight vector at some iteration $t$ is $\mathbf{w}^t$ . The next data-point picked up by the perceptron algorithm in the course of its execution is $(\mathbf{x}, y)$ , where $y$ is the true label:

$\mathbf{w}^t = \begin{bmatrix} 1 \\ 2 \\ 3 \\ 4 \end{bmatrix}, \quad \mathbf{x} = \begin{bmatrix} -1 \\ 0 \\ 1 \\ -1 \end{bmatrix}, \quad y = 1$

What is the value of $\mathbf{w}^{t+1}$ ?

Show answer

Correct answer

Question 5 on its own page

Question 6

+5 marksOne correct option

A
w shown in this diagram is the optimal weight vector for a hard-margin, linear- SVM
B
w shown in this diagram is not the optimal weight vector for a hard-margin, linear-SVM

Show answer

Correct answer

B
w shown in this diagram is not the optimal weight vector for a hard-margin, linear-SVM

Question 6 on its own page

Question 7

+5 marksOne correct option

A
50
B
20
C
30
D
10

Show answer

Correct answer

A
50

Question 7 on its own page

Question 8

+5 marksOne or more correct options

Select all that apply.

Show answer

Correct answers

Question 8 on its own page

Question 9

+5 marksOne or more correct options

Select all that apply.

Show answer

Correct answers

Question 9 on its own page

Question 10

+5 marksOne or more correct options

Select all that apply.

A
80
B
20
C
70
D
30
E
50
F
100

Show answer

Correct answers

A
80
B
20

Question 10 on its own page

Question 11

+5 marksOne or more correct options

Select all that apply.

A
(1) is the loss on the validation dataset
B
(2) is the loss on the training dataset
C
(1) is the loss on the training dataset
D
(2) is the loss on the validation dataset

Show answer

Correct answers

A
(1) is the loss on the validation dataset
B
(2) is the loss on the training dataset

Question 11 on its own page

Question 12

+3 marksOne or more correct options

Select all that apply.

Show answer

Correct answers

Question 12 on its own page

Question 13

+3 marksOne or more correct options

Select all that apply.

Show answer

Correct answers

Question 13 on its own page

Question 14

+5 marksNumerical answer

Show answer

Correct answer: 2.5

Question 14 on its own page

Question 15

+5 marksNumerical answer

Show answer

Correct answer: 0

Question 15 on its own page

Question 16

+5 marksNumerical answer

Consider a binary classification problem for a linearly separable dataset in $\mathbb{R}^4$ . The optimal weight vector for a hard-margin linear-SVM classifier is given to be $\mathbf{w}^*$ . The data-point $(\mathbf{x}, y)$ belongs to the training dataset:

$\mathbf{w}^* = \begin{bmatrix} 1 \\ 0 \\ -1 \\ 2 \end{bmatrix}, \quad \mathbf{x} = \begin{bmatrix} 1 \\ -5 \\ -3 \\ 0 \end{bmatrix}, \quad y = 1$

What is the value of $\alpha^*$ corresponding to this data-point? If you think the answer cannot be determined with this information, enter $-1$ . If you think it can be determined, enter the correct value of $\alpha^*$ . Note that $\alpha^*$ is the Lagrange multiplier corresponding to this data-point.

Show answer

Correct answer: 0

Question 16 on its own page

Question 17

+5 marksNumerical answer

Consider the following architecture for a neural network:

Layer	Neurons
Input	10
Hidden Layer-1	20
Hidden layer-2	30
Output layer	1

How many weights does this network have? Assume that there is no bias associated with any neuron.

Show answer

Correct answer: 830

Question 17 on its own page

Question 18

+5 marksNumerical answer

Show answer

Correct answer: 0.69 (accepted within ±0.01)

Question 18 on its own page

Question 19

+2 marksNumerical answer

Based on the above data, answer the given subquestions.

How many parameters need to be estimated to make a prediction for a data point using a naive Bayes algorithm?

Show answer

Correct answer: 11

Question 19 on its own page

Question 20

+3 marksNumerical answer

Based on the above data, answer the given subquestions.

With what probability does the first feature of a point take the value 0 given that the point is labeled 1? Enter your answer correct to two decimal places.

Show answer

Correct answer: 0.66 (accepted within ±0.02)

Question 20 on its own page

Question 21

+3 marksNumerical answer

Based on the above data, answer the given subquestions.

What will be the misclassification error incurred by the first decision stump?

Show answer

Correct answer: 0.2

Question 21 on its own page

Question 22

+3 marksOne or more correct options

Based on the above data, answer the given subquestions.

To train the second decision stump, which pair of points will be assigned equal weights to create the training dataset?

Select all that apply.

Show answer

Correct answers

Question 22 on its own page

Question 23

+3 marksNumerical answer

Based on the above data, answer the given subquestions.

Show answer

Correct answer: 0.13 (accepted within ±0.02)

Question 23 on its own page