School of Technology, Design and Computer Application

College of Technology
Bachelor of Technology (Information Technology)

Academic
Semester: 6 2024-25
Year:
Artificial Intelligence with
Course Name: concepts of Machine Course Code: 1010103417
Learning & Deep Learning

​ ​ ​ ​ ​ Question bank

Sr.
Question Text Marks CO
No.
Unit 1 : Introduction to AI
1 Define Artificial Intelligence (AI) and its basic concepts. 3 CO1
What are some common tools and technologies used in AI
2 3 CO1
development?
3 Compare AI with traditional programming techniques. 3 CO1
Discuss the characteristics of AI problems. Provide examples to
4 4 CO1
illustrate.
What are the major application areas of AI? Mention at least
5 4 CO1
three with examples.
6 Explain the basic problem-solving methods used in AI. 4 CO1
Compare AI with other emerging technologies, such as machine
7 learning and data science, highlighting their overlaps and 8 CO1
differences.
Discuss the characteristics of AI problems in detail, and explain
8 8 CO1
why these problems are suitable for AI solutions.
 Explain the different application areas of AI, such as healthcare,
9 robotics, and natural language processing, with real-world 8 CO1
examples.
Unit 2 : Introduction to Programming Tools for Data Science
Name three commonly used Python toolkits for data science and
10 3 CO1
their primary applications.
11 What is the purpose of dimensionality reduction in data science? 3 CO1
Explain the difference between bar charts and line charts with an
12 3 CO1
example.
Discuss how APIs can be used to collect data, with an example
13 4 CO1
of the Twitter API.
What are the steps involved in cleaning and munging data?
14 4 CO1
Provide a brief explanation.
Explain how NumPy can be used for manipulating and rescaling
15 4 CO1
data.
Discuss the role of Python libraries such as Matplotlib,
16 Scikit-learn, and NLTK in data science. Provide examples of 8 CO1
tasks they are suited for.
Explain the process of reading files, web scraping, and using
17 APIs to collect and prepare data for analysis. Provide an example 8 CO1
of each.
Describe the methods for visualizing data, such as bar charts,
18 line charts, and scatter plots, and their applications in data 8 CO1
analysis.
Unit 3: Machine Learning
Define overfitting in machine learning and explain how train/test
19 3 CO2
splits can help mitigate it.
Differentiate between supervised and unsupervised learning with
20 3 CO2
examples.
What is Bayes’ theorem? Briefly explain its significance in
21 3 CO2
machine learning.
Compare and contrast classification and regression in the context
22 4 CO2
of machine learning.
Discuss the applications of unsupervised learning in real-world
23 4 CO2
scenarios.
What are evaluation matrices? List and explain metrics for
24 4 CO2
classification algorithms.
Explain the working principle of the K-Nearest Neighbors
25 8 CO2
(KNN) algorithm.
 Explain linear regression and its assumptions. How is it used for
26 8 CO2
predictive analysis?
Discuss the working mechanism of Random Forest. How does it
27 8 CO2
overcome the limitations of Decision Trees?
Unit 4: Deep Learning
What is the difference between biological neural networks and
28 3 CO2
artificial neural networks (ANNs)?
29 Define training and validation in the context of deep learning. 3 CO2
What is Maximum Likelihood Estimation (MLE) in parameter
30 3 CO2
estimation?
Explain the role of activation functions in artificial neural
31 4 CO2
networks.
Compare and contrast Maximum Likelihood Estimation (MLE)
32 4 CO2
and Maximum A Posteriori (MAP) methods.
What are some common challenges faced during the training of
33 4 CO2
neural networks?
Discuss the training process of artificial neural networks,
34 including forward propagation, loss calculation, and 8 CO2
backpropagation.
Explain various network topologies in deep learning with
35 examples, such as feedforward, convolutional, and recurrent 8 CO2
networks.
Describe the techniques for parameter estimation in neural
36 networks, focusing on MLE and MAP. Highlight their 8 CO2
applications and limitations.
Unit 5: Types of NN
What is the primary difference between feedforward neural
37 3 CO3
networks (FNN) and recurrent neural networks (RNN)?
Define Long Short-Term Memory (LSTM) networks and
38 3 CO3
mention one application where they are useful.
Explain how LSTM networks differ from standard RNNs and
39 3 CO3
their advantage in sequence modeling.
Compare and contrast Convolutional Neural Networks (CNN)
40 4 CO3
and Recurrent Neural Networks (RNN).
What are precision and recall? How are they used to evaluate the
41 4 CO3
performance of a model?
42 Discuss the architecture and purpose of LSTM. 4 CO3
Describe the basic structure of a neural network, including its
43 8 CO3
components and how it functions.
 Explain in detail the training process of a neural network,
44 8 CO3
including the role of backpropagation and gradient descent.
List the activation functions in neural networks and explain the
45 8 CO3
role of activation functions in neural networks.
Unit 6: Convolutional Neural Networks (CNNs) Architectures
What are Convolutional Neural Networks (CNNs), and why are
46 3 CO3
they widely used in computer vision tasks?
Define Generative Adversarial Networks (GANs) and explain
47 3 CO3
their basic working principle.
What is the key architectural innovation introduced by ResNet
48 3 CO3
compared to earlier CNN architectures?
Compare LeNet and AlexNet in terms of their architecture and
49 4 CO3
use cases.
What are the main types of GANs, and how do they differ from
50 4 CO3
each other?
Describe the contributions of VGGNet in improving CNN
51 4 CO3
performance and architecture.
Discuss the evolution of CNN architectures, focusing on LeNet,
52 AlexNet, and GoogLeNet, highlighting their key differences and 8 CO3
improvements.
Explain the architecture and functionality of ResNet. How does
53 it use residual connections to address the vanishing gradient 8 CO3
problem?
What are the applications of GANs in real-world scenarios?
54 Explain with examples, including image generation, style 8 CO3
transfer, and data augmentation.

Course Coordinator​ ​ ​ ​ ​ ​ ​ Head of Department​