10/31/24, 8:50 PM Deep Learning - IIT Ropar - - Unit 6 - Week 3

(https://swayam.gov.in) (https://swayam.gov.in/nc_details/NPTEL)

[email protected] 

NPTEL (https://swayam.gov.in/explorer?ncCode=NPTEL) » Deep Learning - IIT Ropar (course)

Course Week 3 : Assignment 3

outline The due date for submitting this assignment has passed.
Due on 2024-08-14, 23:59 IST.
About
NPTEL ()
Assignment submitted on 2024-08-12, 09:39 IST
How does an Use the following data to answer the questions 1 to 3
NPTEL The diagram below shows a neural network. The network contains two hidden layers and one
online output layer. The input to the network is a column vector x ∈ R3 . The first hidden layer contains
course 6 neurons, the second hidden layer contains 4 neurons and the output layer contains 2 neurons.
work? ()
Each neuron in the l th layer is connected to all the neurons in the (l + 1)th layer. Each neuron
has a bias connected to it (not explicitly shown in the figure).
Week 1 ()

Week 2 ()

Week 3 ()

Feedforward
Neural
Networks
(a.k.a
multilayered
network of
neurons)
(unit?
unit=46&lesso
n=47)
In the diagram, W1 is a matrix and x, a1 , h 1 , and O are all column vectors. The notation
Learning Wi [j, :] denotes the jth row of the matrix Wi , Wi [:, j] denotes the jth column of the matrix
Paramters of and Wk denotes an element at ith row and jth column of the matrix Wk .
Wi
ij
Feedforward
Neural 1) Choose the correct dimensions of W1 and a1 1 point
Networks
(Intuition)

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=46&assessment=287 1/4
10/31/24, 8:50 PM Deep Learning - IIT Ropar - - Unit 6 - Week 3

(unit? W1 ∈ R
6×3

unit=46&lesso
n=48) W1 ∈ R
3×6

Output 1×6
W1 ∈ R
functions and
Loss functions
6×1
a1 ∈ R
(unit?
unit=46&lesso
1×6
a1 ∈ R
n=49)

Backpropagati a1 ∈ R
1×1

on (Intuition)
Yes, the answer is correct.
(unit?
Score: 1
unit=46&lesso
Accepted Answers:
n=50) 6×3
W1 ∈ R
6×1
Backpropagati a1 ∈ R

on: Computing
2) How many learnable parameters(including bias) are there in the network?
Gradients w.r.t.
the Output
62
Units (unit?
unit=46&lesso Yes, the answer is correct.
n=51) Score: 1
Accepted Answers:
Backpropagati
(Type: Numeric) 62
on: Computing
Gradients w.r.t. 1 point
Hidden Units
(unit? 3) Which of the following loss functions can be used for the classification problem? 1 point
unit=46&lesso
n=52) Means Squared error
Cross entropy
Backpropagati
on: Computing Both Mean Squared Error and Cross Entropy loss can be used
Gradients w.r.t. No, the answer is incorrect.
Parameters Score: 0
(unit? Accepted Answers:
unit=46&lesso Both Mean Squared Error and Cross Entropy loss can be used
n=53)

Backpropagati 4) We have a multi-classification problem that we decide to solve by training a 1 point

on: Pseudo feedforward neural network. What activation function should we use in the output layer to get the
code (unit? best results?
unit=46&lesso
n=54) Logistic

Derivative of Step function

the activation Softmax
function (unit?
linear
unit=46&lesso
n=55) Yes, the answer is correct.
Score: 1
Information Accepted Answers:
content, Softmax
Entropy &
cross entropy
5) We have data x with the following labels y = [‘car’, ‘bus’, ‘bike’, ‘car’, ‘bus’, ‘car’, 1 point
(unit?
‘bus’, ‘bike’, ‘car’]. Which of the following distribution will give the lowest cross-entropy loss with
y? (Distribution is given in the following order [‘car’, ‘bus’,‘bike’])?

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=46&assessment=287 2/4
10/31/24, 8:50 PM Deep Learning - IIT Ropar - - Unit 6 - Week 3

unit=46&lesso
n=56) [0.49, 0.30, 0.21]

Lecture
[0.44, 0.33, 0.22]
Material for
Week 3 (unit?
[0.50, 0.30, 0.20]
unit=46&lesso
n=57)
[0.44, 0.39, 0.19]

Quiz: Week 3 No, the answer is incorrect.

: Assignment Score: 0
3 Accepted Answers:
(assessment? [0.44, 0.33, 0.22]

name=287)
6) Which of the following statements about backpropagation is true? 1 point
Week 3
Feedback It is used to compute the output of a neural network.
Form: Deep
It is used to initialize the weights in a neural network.
Learning - IIT
Ropar (unit?
It is used to optimize the weights in a neural network.
unit=46&lesso It is used to regularize the weights in a neural network.
n=186)
Yes, the answer is correct.
Score: 1
week 4 ()
Accepted Answers:
It is used to optimize the weights in a neural network.
Week 5 ()

7) Given two probability distributions p and q , under what conditions is the cross 1 point
Week 6 ()
entropy between them minimized?

Week 7 ()
All the values in p are lower than corresponding values in q
Week 8 ()
All the values in p are lower than corresponding values in q
Week 9 ()
p = 0 (0 is a vector)

week 10 () p = q

Yes, the answer is correct.

Week 11 () Score: 1
Accepted Answers:
Week 12 () p = q

Download 8) Given that the probability of Event A occurring is 0.80 and the probability of Event B 1 point
Videos () occurring is 0.20, which of the following statements is correct?

Event A has a low information content

Books ()
Event A has a high information content
Text Event B has a low information content
Transcripts Event B has a high information content
()
Yes, the answer is correct.
Score: 1
Problem
Accepted Answers:
Solving Event A has a low information content
Session -
Event B has a high information content
July 2024 ()

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=46&assessment=287 3/4
10/31/24, 8:50 PM Deep Learning - IIT Ropar - - Unit 6 - Week 3

Use the following data to answer the questions 9 and 10The following diagram represents a
neural network containing two hidden layers and one output layer. The input to the network is a
column vector x ∈ R
3
. The activation function used in hidden layers is sigmoid. The output layer
doesn't contain any activation function and the loss used is squared error loss
(predy − truey )
2
.

The following network doesn't contain any biases and the weights of the network are given
below:
1 1 3
⎡ ⎤
1 1 2
W 1 =⎢ 2 1 1 ⎥ W 2 =[ ] W 3 =[ 2 3]
3 2 1
⎣ ⎦
1 2 3

1
⎡ ⎤

The input to the network is: x =⎢ 1 ⎥

⎣ ⎦
1

The target value y is: y = 8

9) What is the predicted output for the given input x after doing the forward pass? (Choose the
option closest to your answer)

4.955

Yes, the answer is correct.

Score: 1
Accepted Answers:
(Type: Range) 4.7,5.2

1 point

10) Compute and enter the loss between the output generated by input x and the true output y

9.272

Yes, the answer is correct.

Score: 1
Accepted Answers:
(Type: Range) 8.82,9.74

1 point

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=46&assessment=287 4/4