10/24/24, 10:08 AM Deep Learning - IIT Ropar - - Unit 9 - Week 6

(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 6 : Assignment 6

outline The due date for submitting this assignment has passed.
Due on 2024-09-04, 23:59 IST.
About
NPTEL ()
Assignment submitted on 2024-09-04, 13:46 IST
How does an
NPTEL
online
course
work? ()

Week 1 ()

Week 2 ()

Week 3 ()

week 4 ()

Week 5 ()

Week 6 ()

Introduction to
Autoncoders
(unit?
unit=83&lesso
n=84)

Link between
PCA and
Autoencoders
(unit?

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=83&assessment=294 1/5
10/24/24, 10:08 AM Deep Learning - IIT Ropar - - Unit 9 - Week 6

unit=83&lesso 1) Which of the following networks represents an autoencoder? 1 point

n=85)

Regularization
in
autoencoders
(Motivation)
(unit?
unit=83&lesso
n=86)

Denoising
Autoencoders
(unit?
unit=83&lesso
n=87)

Sparse
Autoencoders
(unit?
unit=83&lesso
n=88)

Contractive
Autoencoders
(unit?
unit=83&lesso
n=89)

Lecture
Material for
Week 6 (unit?
unit=83&lesso
n=90)

Quiz: Week 6
: Assignment
6
(assessment?
name=294)

Week 6
Feedback
Form: Deep
Learning - IIT
Ropar (unit?
unit=83&lesso
n=235) a

Week 7 () b
c
Week 8 () d

No, the answer is incorrect.

Week 9 () Score: 0
Accepted Answers:
week 10 () b

Week 11 ()

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=83&assessment=294 2/5
10/24/24, 10:08 AM Deep Learning - IIT Ropar - - Unit 9 - Week 6

2) If an under-complete autoencoder has an input layer with a dimension of 7, what 1 point

Week 12 ()
could be the possible dimension of the hidden layer?

Download 6
Videos ()
8

Books () 0
7
Text 2
Transcripts
No, the answer is incorrect.
() Score: 0
Accepted Answers:
Problem 6
Solving 2
Session -
July 2024 ()
3) What is the primary reason for adding corruption to the input data in a denoising 1 point
autoencoder?

To increase the complexity of the model.

To improve the model’s ability to generalize to unseen data.
To reduce the size of the training dataset.
To increase the training time.

Yes, the answer is correct.

Score: 1
Accepted Answers:
To improve the model’s ability to generalize to unseen data.

4) What is/are the primary advantages of Autoencoders over PCA? 1 point

Autoencoders are less prone to overfitting than PCA.

Autoencoders are faster and more efficient than PCA.
Autoencoders can capture nonlinear relationships in the input data.
Autoencoders require fewer input data than PCA.

No, the answer is incorrect.

Score: 0
Accepted Answers:
Autoencoders can capture nonlinear relationships in the input data.

5) What type of autoencoder is it when the hidden layer’s dimensionality is less than 1 point
that of the input layer?

Under-complete autoencoder
Complete autoencoder
Overcomplete autoencoder
Sparse autoencoder

No, the answer is incorrect.

Score: 0
Accepted Answers:
Under-complete autoencoder

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=83&assessment=294 3/5
10/24/24, 10:08 AM Deep Learning - IIT Ropar - - Unit 9 - Week 6

6) Which of the following statements about overfitting in overcomplete autoencoders is 1 point

true?

Reconstruction error is very low while training

Reconstruction error is very high while training
Network fails to learn good representations of input
Network learns good representations of input
Yes, the answer is correct.
Score: 1
Accepted Answers:
Reconstruction error is very low while training
Network fails to learn good representations of input

7) What is the purpose of a decoder in an autoencoder? 1 point

To reconstruct the input data

To generate new data
To compress the input data
To extract features from the input data

No, the answer is incorrect.

Score: 0
Accepted Answers:
To reconstruct the input data

8) What are the advantages of using a denoising autoencoder? 1 point

Robustness to noisy input data

Reduction of the risk of overfitting
Faster training time
It promotes sparsity in the hidden layer

Yes, the answer is correct.

Score: 1
Accepted Answers:
Robustness to noisy input data
Reduction of the risk of overfitting

9) What is the primary objective of sparse autoencoders that distinguishes it from 1 point
vanilla autoencoder?

They learn a low-dimensional representation of the input data

They minimize the reconstruction error between the input and the output
They capture only the important variations/features in the data
They maximize the mutual information between the input and the output

No, the answer is incorrect.

Score: 0
Accepted Answers:
They capture only the important variations/features in the data

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=83&assessment=294 4/5
10/24/24, 10:08 AM Deep Learning - IIT Ropar - - Unit 9 - Week 6

10) If the dimension of the hidden layer representation is more than the dimension of 1 point
the input layer, then what kind of autoencoder do we have?

Complete autoencoder
Under-complete autoencoder
Overcomplete autoencoder
Sparse autoencoder
Yes, the answer is correct.
Score: 1
Accepted Answers:
Overcomplete autoencoder

https://onlinecourses.nptel.ac.in/noc24_cs114/unit?unit=83&assessment=294 5/5