Scribd
Upload
25 views2 pages

CST414 A

This document outlines the examination structure for the Eighth Semester B.Tech Degree Regular Examination in Deep Learning at APJ Abdul Kalam Technological University. It includes two parts: Part A with ten short answer questions and Part B with one full question from each of the five modules, covering various topics in deep learning. The exam is scheduled for June 2023, with a maximum score of 100 marks and a duration of 3 hours.

Uploaded by

devanand272003
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
25 views2 pages

CST414 A

This document outlines the examination structure for the Eighth Semester B.Tech Degree Regular Examination in Deep Learning at APJ Abdul Kalam Technological University. It includes two parts: Part A with ten short answer questions and Part B with one full question from each of the five modules, covering various topics in deep learning. The exam is scheduled for June 2023, with a maximum score of 100 marks and a duration of 3 hours.

Uploaded by

devanand272003
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

B 0400CST414052301 Pages: 2

Reg No.:_______________ Name:__________________________


APJ ABDUL KALAM TECHNOLOGICAL UNIVERSITY
Eighth Semester B.Tech Degree Regular Examination June 2023 (2019 Scheme)

Course Code: CST414


Course Name: DEEP LEARNING
Max. Marks: 100 Duration: 3 Hours

PART A
Answer all questions, each carries 3 marks. Marks

1 Discuss the disadvantages of single layer perceptrons with an example. (3)


2 List any three applications of neural network. (3)
3 Explain any two parameter initialization methods. (3)
4 Compare L1 and L2 regularization. (3)
5 Define structured outputs in Convolutional Neural networks. (3)
6 Suggest a method to make convolution algorithm more efficient. Justify Your (3)
answer.
7 Explain Recursive Neural Network. (3)
8 Explain applications of Recurrent Neural Network. (3)
9 What is an autoencoder? Give one application of an autoencoder. (3)
10 Illustrate the use of deep learning concepts in Computer Vision. (3)
PART B
Answer any one full question from each module, each carries 14 marks.
Module I
11 a) Explain different activations functions and their derivatives used in neural (14)
networks with the help of graphical representation.
OR
12 a) Implement the back propagation algorithm to train Multi Layer Perceptron using (14)
tanh activation function.
Module II
13 a) Explain different gradient descent optimization strategies used in deep learning. (14)
OR
14 a) Explain the following. i)Early stopping ii) Drop out iii) Injecting noise at input (14)
iv)Parameter sharing and tying.

Page 1of 2
0400CST414052301

Module III
15 a) Explain variants of convolution functions in detail. (14)
OR
16 a) Sketch the diagram of Convolutional Neural Network architecture and explain (14)
different stages in detail.
Module IV
17 a) Sketch diagrams of different Recurrent Neural Network patterns and explain them (14)
in detail.
OR
18 a) Discuss the architecture of Gated Recurrent Unit (GRU). (7)
b) Discuss different ways to make a Recurrent Neural Network(RNN) deep RNN (7)
with the help of diagrams.
Module V
19 a) Illustrate the use of representation learning in object classification. (7)

b) Compare Boltzmann Machine with Deep Belief Network. (7)


OR
20 a) Explain any three research areas of neural network. (14)
****

Page 2of 2

You might also like