Scribd
Upload
20 views2 pages

Btech Cse 7 Sem 8 Sem Deep Learning 90495 Dec 2022

This document outlines the examination details for a B.Tech. (CSE) course on Deep Learning, including instructions for candidates and the structure of the exam divided into three sections. Section-A consists of ten compulsory questions worth two marks each, Section-B has five questions with four to be attempted worth five marks each, and Section-C includes three questions with two to be attempted worth ten marks each. The document also includes specific topics and concepts related to deep learning that students are expected to address in their answers.

Uploaded by

shailendrakumar9519842594
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views2 pages

Btech Cse 7 Sem 8 Sem Deep Learning 90495 Dec 2022

This document outlines the examination details for a B.Tech. (CSE) course on Deep Learning, including instructions for candidates and the structure of the exam divided into three sections. Section-A consists of ten compulsory questions worth two marks each, Section-B has five questions with four to be attempted worth five marks each, and Section-C includes three questions with two to be attempted worth ten marks each. The document also includes specific topics and concepts related to deep learning that students are expected to address in their answers.

Uploaded by

shailendrakumar9519842594
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Roll No. Total No.

of Pages : 02
Total No. of Questions : 09
B.Tech. (CSE) (Sem.–7/8)
DEEP LEARNING
Subject Code : BTCS-704-18
M.Code : 90495
Date of Examination : 23-12-22
Time : 3 Hrs. Max. Marks : 60

INSTRUCTIONS TO CANDIDATES :
1. SECTION-A is COMPULSORY consisting of TEN questions carrying T WO marks
each.
2. SECTION-B contains FIVE questions carrying FIVE marks each and students
have to attempt any FOUR questions.
3. SECTION-C contains T HREE questions carrying T EN marks each and students
have to attempt any T WO questions.

SECTION-A

1. Write briefly :

a) Define the term ‘deep learning’.

b) Differentiate between bias and variance.

c) Explain briefly about estimators.

d) What ‘‘is’’ the advantage associated with using computational graphs?

e) What is dropout rate?

f) Define the term padding in convolutional neural networks.

g) Differentiate between Recursive neural networks and Echo state networks.

h) What is Sequence modelling?

i) Explain briefly about a directed generative network?

j) Discuss in brief about Sigmoid belief networks.

1 | M- (S2)-1235
SECTION-B

2. What is an activation function? Explain about various activation functions.

3. Explain in detail about the working of a sequence-to-sequence model.

4. What is a convolution operation? Explain about neuroscientific basis for convolutional


networks.

5. What are gradients? How are they useful in finding an optimal value of the parameters?

6. What is an auto-encoder? Explain various components of an auto-encoder along with


the properties of auto-encoders.

SECTION-C

7. Explain the working of back propagation algorithm with the help of an example.

8. Why are convolutional neural networks preferred over feed-forward neural networks
especially in the case of images? Also explain about the role of ‘kernel’ and ‘stride’ in the
convolution process.

9. Discuss in detail about various similarities and differences between Deep Belief
Networks and Deep Boltzmann machines.

NOTE : Disclosure of Identity by writing Mobile No. or Making of passing request on any
page of Answer Sheet will lead to UMC against the Student.

2 | M- (S2)-1235

You might also like