School of Engineering (ML-DL)I

Year M. Tech - II Semester

Subject Name: Data Warehouse & Data Mining
Subject Code: MR22-2CS0203
Question Bank

UNIT-I
1. Differentiate Database system and Data warehouse system?
2. Explain in Detail Multi-dimensional Data Model?
3. Explain the Three tier Data warehousing in detail with suitable example?
4. Explain in detail the following
I. Stars Flakes
II. Snow Flakes,
III. Fact Constellations
5. Explain in detail the various data mart with suitable example?
UNIT-2
1. Illustrate in detail about Decision tree in detail with suitable example?
2. Explain K-nearest neighbor classifier Algorithm in detail with suitable example?
3. Explain Genetic Algorithm in detail with suitable example?
4. List out different types of Categories of clustering methods and explain about tree based
clustering methods?
5. Explain in detail about Density Based Methods (DBSCAN, OPTICS) and Grid Based Methods
(STING, CLIQUE)?
UNIT-3
1. With suitable example explain Apriori in algorithm in detail?
2. Explain Data Generalization in detail with suitable example?
3. Explain the various Basic Statistical Graph class Description in detail with suitable?
4. Briefly explain about Multi-Dimensional association rule mining in Transactional databases?
5. Explain the various measures that can be adopted for measuring Central Tendency?
UNIT-4
1. Explain the various pre-processing in detail with suitable example?
2. Explain in detail about the Data Cube Aggregation?
3. Explain Concept hierarchy generation in detail with suitable example?
4. Discus the various Dimensionality reduction techniques in detail?
5. Explain the Data compression techniques in detail?
UNIT-5
1. Describe about different OLAP servers in detail?
2. Explain in detail about Backup and recovery operations in Data warehousing?
3. Explain the Data Aggregation in detail with suitable example?
4. Explain about various Data mining interface in detail?
5. Explain the various Testing techniques in Data Warehousing ?
 SchoolofEngineering (ML-DL)
I Year M.Tech-II Semester
Subject Name:Deep Learning
Subject Code: MR22-2CS0206
Question Bank
Unit1
1. How did the historical development of deep learning lead to the creation of Multilayer
Perceptrons (MLPs), and how do MLPs differ from earlier models like the Perceptron?
2. Explain the significance of the McCulloch-Pitts neuron in the context of the history of
deep learning and its role in shaping the development of artificial neural networks?
3. What is the Perceptron Learning Algorithm, and how did its introduction mark a
significant step forward in the evolution of neural network models, particularly in terms
of pattern recognition and classification tasks?
4. In the context of feedforward Neural Networks, could you elaborate on the representation
power they offer and how they overcome the limitations of single-layer models like the
Perceptron?
5. Write comprehensive overview of the Backpropagation algorithm, its role in training
deep networks like Multilayer Perceptrons, and how it addresses the challenges of
adjusting weights in complex architectures?
Unit 2
1. How does the Momentum-based Gradient Descent algorithm differ from traditional
Gradient Descent, and how does it address challenges like oscillations and slow
convergence by incorporating the concept of velocity?
2. compare the Nesterov Accelerated Gradient Descent with regular Momentum-based
Gradient Descent, explaining how the former's "look-ahead" mechanism improves
convergence rates and overall performance in optimizing complex loss landscapes?
3. What are the fundamental differences between Stochastic Gradient Descent and
traditional Gradient Descent, and how does the introduction of randomness impact
convergence speed and exploration of parameter space?
4. Elucidate the key concepts behind optimization techniques like AdaGrad, RMSProp, and
Adam, highlighting their unique adaptations of learning rates and how they address issues
such as vanishing and exploding gradients in neural network training?
5. How do eigenvalues and eigenvectors play a crucial role in the Eigenvalue
Decomposition process, and how can this decomposition be utilized for tasks like
dimensionality reduction, as seen in Principal Component Analysis (PCA)? Furthermore,
what are the practical implications and interpretations of the principal components
extracted through PCA?

Unit 3
1. Explain the connection between Autoencoders and Principal Component Analysis (PCA),
detailing how the encoding and decoding process of autoencoders relates to the extraction
of principal components in PCA, and how they differ in handling nonlinear relationships?
2. How do Denoising Autoencoders address the challenge of noisy input data, and what are
the underlying mechanisms that enable them to reconstruct clean representations while
simultaneously learning useful features from corrupted input?
3. Elaborate on the concept of sparse autoencoders and their significance in feature learning,
highlighting how their enforced sparsity constraint contributes to the discovery of more
informative and compact representations within the encoded data?

4. What role do Contractive Autoencoders play in learning invariant representations? How

do they impose constraints on the encoder to create robust features that are insensitive to
small perturbations in the input data?
5. Discuss the Bias-Variance tradeoff and how regularization techniques like L2
regularization, early stopping, and dropout contribute to managing this tradeoff in
machine learning models, including autoencoders. How do these methods impact the
model's generalization and ability to handle unseen data?
 Unit 4
1. What is the principle behind Greedy Layer-wise Pre-training in deep neural networks,
and how does it contribute to overcoming the challenges of training very deep
architectures? Can you provide examples of how this technique has been used effectively
in real-world applications?
2. How do modern activation functions, such as ReLU (Rectified Linear Unit) and its
variants, address the limitations of traditional sigmoid and tanh functions? Could you
elaborate on how these new activation functions improve convergence, mitigate
vanishing gradients, and accelerate training in deep networks?
3. Explain the significance of better weight initialization methods, like He initialization and
Xavier initialization, in deep neural networks? How do these techniques contribute to
reducing the vanishing and exploding gradient problems and improving the overall
training process?
4. What role does Batch Normalization play in the training of deep neural networks? How
does it mitigate internal covariate shift, and what are the benefits in terms of faster
convergence, improved generalization, and more stable training dynamics?
5. write overview of Word2Vec, a widely used technique for learning vector representations
of words from large text corpora? How do the Continuous Bag of Words (CBOW) and
Skip-gram models work, and what are the practical applications of these learned word
embeddings in natural language processing tasks?
Unit5
1. What are the key architectural features that differentiate Convolutional Neural Networks
(CNNs) from traditional fully connected neural networks? How do CNNs exploit spatial
hierarchies in data and achieve translational invariance, making them particularly
effective for image recognition tasks?
2. Recurrent Neural Networks (RNNs) are known for their ability to process sequential data.
How does Backpropagation through time (BPTT) work in training RNNs, and what
challenges does it pose in terms of vanishing and exploding gradients over long
sequences?
3. Compare and contrast Gated Recurrent Units (GRUs) and Long Short-Term Memory
(LSTM) cells, two popular types of recurrent layers. How do they address the vanishing
gradient problem and enable RNNs to capture long-range dependencies more effectively?
4. What is the Attention Mechanism in neural networks, and how does it enhance the
performance of sequence-to-sequence tasks, like machine translation? How does attention
extend from sequential data to images, allowing models to focus on relevant regions
within an image for improved understanding?
5. Write a case study analysis of AlexNet, GoogLeNet, and ResNet, outlining their
significant contributions to the field of deep learning and computer vision? How did these
architectures introduce innovations like deeper networks, parallel pathways, and residual
connections to improve accuracy and efficiency in image classification?
 School of Engineering (ML-DL)
I Year M. Tech - II Semester
Subject Name: CLOUD COMPUTING
Subject Code: MR22-2CS0207
Question Bank

UNIT-I

1. Define cloud computing? Draw a neat diagram of cloud computing

architecture.
2. Explain cloud components in detail.
3. Elaborate five essential characteristics of cloud computing with examples.
4. Compare cloud computing and traditional computing.
5. Illustrate roots of Cloud Computing.

UNIT-II
1. Explain the following:
i) High performance computing
ii) Utility computing
iii) Grid computing.

2. Explain technology innovations of cloud computing in detail.

3. What is Cloud Scalability? Elaborate the types of Cloud Scalability.

4. Justify cloud security challenges in detail.

5. Discuss the Regulatory Issues of Cloud Computing and the Government Policies.

UNIT-III

1. Explain in detail about the layered Architecture of Cloud.

2. What are the differences between PaaS, IaaS and SaaS?
3. What is PaaS? Discuss working of Platform as a Service (PaaS) and mention its benefits.
4. Describe the services provided by IaaS provider. Discuss Advantages and disadvantages of
IaaS cloud computing layer.
5. Illustrate various cloud computing deployment models.
 UNIT-IV

1. Explain the layered CloudSim Architecture with a neat diagram.

2. Describe the overall class design diagram of CloudSim and its
importance?
3. Explain the architecture of the Green Cloud Simulation environment?
4. Compare the performance of any three cloud computing simulators?
5. Explain the various cloud simulators and mention its importance?

UNIT-V

1. Define Virtual Machine and mention its applications?

2. Explain the abstract map of VMware Workstation with a neat diagram?
3. Describe the components of VMware infrastructure with suitable
diagram?
4. Explain the operations to be performed in virtual machine of VMware?
5. Describe Physical topology used in VMware infrastructure?
 School of Engineering (ML-DL)
I Year M. Tech - II Semester
Subject Name: Advanced Machine Learning
Subject Code: MR22-2CS0211
Question Bank
UNIT-I
1. What is data mining? Explain the steps in data mining process.
2. Define association rule mining. How are association rules from large
datasets?
3. Define Apriori Algorithm and explain with an example.
4. Write about Introduction and challenges of Data Mining.?
5. Define FP-Growth Algorithm and explain with an example.
UNIT-II
1. Explain about map reduce concepts with an example distributed file systems, complexity theory for
map reduce
2. Discuss about finding similar items using LSH.
3. Explain about Page Ranking of websites with possible iterations.
4. Write in detail about efficient computation of page rank.
UNIT-III
1. Introduction about Review of linear algebra .
2. Explain in detail about Eigen values and Eigen vectors.
3. Explain about Singular Value Decomposition (SVD).
4. Describe the concept of Content-Based Recommendations.
5. Explain in detail about Collaborative Filtering.
UNIT-IV
1. Explain Mining Social Network Graphs Concept in detail.
2. Explain about the Graph Clustering and Graph Partitioning.
3. Explain overlapping communities in detail.
4. Explain about Simrank.
5. Explain about Counting Triangles.
UNIT-V
1. Explain about the Dimensionality Reduction: Principal Component
Analysis (PCA).
2. Describe Feature Selection and Feature Reduction .
3. Explain in detail about Text representations.
4. Explain in detail about naive Bayes and multinomial
Models.
5. Explain in detail about clustering and latent space models.
6. Describe VC- dimension and structural risk minimization, margin
methods
7. Introduction to Deep Learning
 SCHOOL OF ENGINEERING
DEPARTMENT OF AI & ML
QUESTION BANK

SCHOOL OF ENGINEERING
DEPARTMENT OF AI & ML
M.Tech Machine Learning-Deep Learning

Subject Name: Software Engineering Subject Code: MR22-2CS0214

Unit-I

1. Explain in detail the following software process models with a neat diagram
2. Explain the following
(i) Waterfall model
(ii) Spiral model
3. Explain in detail about RAD model with a neat sketch.
4. Discuss the various life cycle models in software development.
5. List the principles of agile software development and explain in detail.

Unit-II

1. Define requirement engineering. State its process and explain requirements elicitation problem.
2. Explain with example diagram the functional and behavioral modeling.
3. Differentiate functional and non-functional requirements.
4. Write an SRS document for Library Management System.
5. Write short notes on requirements elicitation. Explain various activities performed in it.

Unit - III

1. Explain the fundamental software design concepts in detail.

2. List the different types of architectural styles exist for software and explain any one software
architecture in details.
3. a) Explain about the various design concepts considered during design.
b) Write short notes on user interface design process.
4. Discuss in detail about the design process in software development process.
5. Explain in detail about the characteristics and criteria for a good design.

Unit - IV

1. Explain in detail about Class Diagram and Activity Diagram.

2. Define software risks. Elaborate the concepts of Risk management.
3. a)Differentiate between Reactive vs. Proactive Risk strategies.
b) Differentiate between Risk identification and Risk projection.
4. Define RMMM plan. Explain about RMMM Plan in detail.
5. Explain about Quality concepts and Software Quality assurance.

Unit – V

1. Write a note of
(i) Black box testing
(ii) White box testing
2. Define unit testing. Explain its significance. Elaborate the unit testing procedure.
3. Explain the Unit Testing and Integration Testing in detail with suitable example.
4. Explain the Acceptance Testing and Performance Testing in detail with suitable example.
5. Explain the various testing strategies in detail.
 School of Engineering
M.Tech(ML& DL)- I Year- II Semester
Subject Name: Quantum Computing Subject Code: MR22-2BS0131

Question Bank

Unit-I: Introduction to Quantum Computing

1) What is the need of quantum computing and discuss the types of computing.
2) What is the future of quantum computing ?
3) Discuss the applications of quantum computing in artificial intelligence
4) How can quantum computing impact the field of drug discovery and materials design ?
5) Discuss the potential role of quantum computing in accelerating machine learning algorithms.
Unit-II: Quantum Mechanics for Measurements
6) What is complex vector space and discuss its properties.
7) Describe the Dirac notation (Bra-Ket formalism) and its advantages in representing
quantum states.
8) Define Hermitian and unitary operators, and discuss their significance in quantum
mechanics.
9) What are the Pauli operators, and how are they used in quantum computations and
quantum gates?
10) Explain the postulates of quantum mechanics.
Unit-III: Quantum Bits and Quantum Gates
11) (a) What is a Quantum Bit (qubit), and how is it different from a classical bit ?
(b) Explain the concept of superposition in quantum computing and its significance for
qubits.
12) Explain the concept of superposition and entanglemenr in quantum computing and its
significance for qubits.
13) What are quantum gates, and how do they operate on qubits to perform computations?
14) Explain the Hadamard gate and its ability to create superpositions in quantum circuits.
15) Describe the Pauli-X,Y and Z gates and its role in quantum computing. How does it
differ from the classical NOT (CNOT) gate ?
M.Tech (DL& ML) Quantum Computing Page 2 of 2

Unit-IV: Quantum Algorithms

16) Explain the concept of quantum parallelism and how it is utilized in Deutsch’s algorithm.
17) What is the main challenge in classical factoring algorithms, and how does Shor’s
algorithm overcome it?
18) Explain how Grover’s search algorithm achieves the quadratic speedup over classical
search algorithms.
19) What is quantum annealing and how it relates to finding the ground state of a system.
20) Explain quantum noise in quantum computing. How can quantum error correction
codes help preserve the integrity of quantum information in the presence of noise ?

Unit-V: Basic quantum circuit programming using Qiskit

21) What is Qiskit, and what are its main features ?
22) What are quantum circuits, explain implementation of Pauli X gate using Qiskit ?
23) What are quantum circuits, explain implementation of CNOT gate using Qiskit ?
24) Explain implementation of Toffoli Gate using Qiskit.
25) What quantum gates are involved in creating a quantum circuit for the OR gate in
Qiskit

Page 2