Scribd
Upload
3K views2 pages

IF4071 Deep Learning QP

This document is a question paper for the M.E. Computer Science and Engineering program at Anand Institute of Higher Technology, focusing on Deep Learning for the November/December 2024 examinations. It outlines course outcomes, including feature extraction, image segmentation, and traffic information analysis, along with a structured format of questions divided into three parts: Part A, Part B, and Part C. The questions assess various Bloom's levels and course outcomes related to deep learning concepts and applications.

Uploaded by

dhurgadevi.m
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
3K views2 pages

IF4071 Deep Learning QP

This document is a question paper for the M.E. Computer Science and Engineering program at Anand Institute of Higher Technology, focusing on Deep Learning for the November/December 2024 examinations. It outlines course outcomes, including feature extraction, image segmentation, and traffic information analysis, along with a structured format of questions divided into three parts: Part A, Part B, and Part C. The questions assess various Bloom's levels and course outcomes related to deep learning concepts and applications.

Uploaded by

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

Question Paper Code:<Last part of the Order>

Register No:

ANAND INSTITUTE OF HIGHER TECHNOLOGY


An Autonomous Institution, Affiliated to Anna University
CHENNAI -603 103
M.E/M.TECH. DEGREE END SEMESTER EXAMINATIONS NOV/DEC 2024
Third Semester
M.E. COMPUTER SCIENCE AND ENGINEERING
IF4071 -DEEP LEARNING
(Regulation 2021)
Course Outcome:
CO1: Feature Extraction from Image and Video Data
CO2: Implement Image Segmentation and Instance Segmentation in Images
CO3: Implement image recognition and image classification using a pretrained network (Transfer Learning)
CO4: Traffic Information analysis using Twitter Data
CO5: Autoencoder for Classification & Feature Extraction
Bloom’s Level: 1 - Remembering, 2 - Understanding, 3 - Applying, 4 - Analyzing, 5 - Evaluating 6 - Creating.
Answer All the Questions
Bloom’s Course Marks
PART A (10 x2 =20 marks) Level Outcome Allotted

1. Define Probabilistic modeling and state its advantages. [BL1] [CO1] [2]
What are the main differences between AI, Machine Learning, and
2. [BL1] [CO1] [2]
Deep Learning?
3. What is regularization? [BL1] [CO2] [2]
4. List the different Types of Activation Functions [BL1] [CO2] [2]
5. Summarize the Properties of LTI Systems. [BL2] [CO3] [2]
The output layer of convolutional network is usually relatively
6. [BL4] [CO3] [2]
inexpensive to learning layer. Justify
7. Why NLP is difficult? [BL1] [CO4] [2]
8. Define Bag of Words (BOW)and Continuous Bag of Words (CBOW) [BL1] [CO4] [2]
9. Outline the challenges of Deep Reinforcement Learning. [BL2] [CO5] [2]
10. Differentiate Q-Learning and Deep Q-Learning. [BL2] [CO5] [2]
PART B (5 x13 =65 marks)
Explain in detail about the types of perceptron and primary
11.a [BL2] [CO1] [13]
components of a perceptron.
OR
What is Probabilistic Modeling? Explain with examples of Probabilistic
11.b [BL2] [CO1] [13]
Models and state its advantages.
12.a Describe the different types of optimizers and their advantages. [BL2] [CO2] [13]
OR
12.b i. Discuss over fitting and Under fitting with example. [BL2] [CO2] [6]
Question Paper Code:<Last part of the Order>

ii. Analyze how to deal feature engineering. [BL4] [CO2] [7]


What is Convolutional Neural Network? Write the steps in building a
13.a [BL1] [CO3] [13]
CNN training and evaluation of the CNN.
OR
Illustrate about Transfer Learning and when to use Transfer Learning
13.b [BL2] [CO3] [13]
and how implement Transfer Learning?
Explain the Word2Vec with architecture and General Flow of the
14.a [BL2] [CO4] [13]
Algorithm
OR
i. Analyze the working procedure of Gated Recurrent Unit (GRU) with
14.b [BL4] [CO4] [13]
its architecture.
ii. Write about Bi-directional LSTM.
With a neat sketch explain the Generative Adversarial Networks with
15.a [BL2] [CO5] [13]
its components and Steps for Training.
OR
Justify your answer how auto encoders are suitable compared to
15.b [BL5] [CO5] [13]
principal compact analysis (PCA) for dimensionality reduction.
PART C (1 x15 =15 marks)
Develop a Simple Photo Classifier using Inception Oxford VGG
16.a [BL3] [CO3 ] [15]
Model.
OR
With appropriate Pseudo code analyze and explain the working of
16.b [BL4] [CO2] [15]
Actor-Critic Algorithm Policy Gradient Methods.

You might also like