Centre for Artificial Intelligence

Islamic University of Science & Technology

Master of Technology
Artificial Intelligence

Curriculum
for
Master of Technology Artificial Intelligence
Two Year Programme
2024 Onwards
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Programme Description
The Master of Technology (M.Tech) Artificial Intelligence (AI) at the Centre for AI, Islamic University
of Science & Technology, is a two-year program designed to provide students with
comprehensive knowledge and practical skills in the field of AI. This program encompasses a
wide range of topics including mathematics, statistics, programming, machine learning, deep
learning, and their applications. Emphasis is placed on both theoretical foundations and hands-on
experience to prepare students for careers in AI which includes AI in industry and research.

Program Educational Objectives (PEOs)

PEO 1
Focuses on building foundational skills in core areas such as mathematics, statistics, and
programming languages.

PEO 2
Aims to provide deep understanding and hands-on skills in Data Science, Machine Learning,
Computer Vision, and Natural Language Processing.

PEO 3
Concentrates on the use of state-of-the-art tools and high-performance computing platforms for
problem-solving in AI.

PEO 4
Empowers students to tackle AI challenges with a rigorous and analytical mindset, promoting
creativity and interdisciplinary collaboration, aligned with NEP 2020.

PEO 5
Inculcates professionalism and ethical foundations, encouraging contributions to societal
challenges through AI in line with NEP 2020.

PEO 6
Equips students with industry-relevant skills and expertise to meet the evolving demands of the
professional industry.

1
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Intake Capacity, Eligibility Criteria and Mode of Selection

Intake Capacity: 30

Eligibility Criteria
B. Tech./ BE in any branch of Engineering or MCA/ M. Sc. Information Technology/ M. Sc.
Electronics or equivalent with a minimum of 60% marks.

Mode of Selection
Candidates will be primarily selected through their performance in the GATE examination. Should
there remain unfilled seats, the university will conduct its own entrance exam to select additional
candidates.

Program Structure
The proposed M. Tech program would span for two years. The course curriculum is developed in
accordance with the AICTE Model Curriculum in consultation with industry experts. It shall be
regularly updated to align with the latest advancements and industry requirements. It comprises
core concepts, practical skills, and hands-on experience in a balanced blend of theoretical
coursework, laboratory exercises, and industry-oriented projects.

A. Definition of Credit

Credit is one of the primary methods used to determine and document that student has
met academic requirements. Credits are awarded upon completing and passing a course.

1 hr. Lecture (L) per week 1 credit

1 hr. Tutorial (T) per week 1 credit

2 hrs. Practical (P) per week 1 credit

B. Range and distribution of Credits

Credits earned in the range of 77 and above shall be required for a student to be eligible
to get MTech Artificial Intelligence.

2
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Distribution of Credits

S No Semester Total Credits

1 I 23

2 II 20

3 III 20

4 IV 14

Total Credits 77

C. Course Code and Definition

All courses (except Open Electives) are denoted by a seven-digit alphanumeric code
(XXXXXXX), three alphabets followed by three numerals, followed by one alphabet .
1. The first three alphabets designate the department teaching the course, i.e., the
discipline to which the course belongs, e.g., CAI for Centre for Artificial
Intelligence.
2. The first numeral following the three alphabets indicate the level of the course, 1 to
4 for undergraduate 1st to 4th year; 5 to 7 for postgraduate 1st to 3rd year, and 8, 9
for PhD.
3. The next two numerals are the unique identification numbers for the course.
Courses running in odd semesters are labeled from 01 to 49 and courses running
in even semesters are labeled from 50 to 99.
4. The last alphabet indicates the nature of the course. It is one amongst four
choices, C (Core Course), E (Elective (Discipline Centric)), G (Elective (Generic)), F
(Foundation Course). Since, open electives are identified by a zero in place of the
level numeral (at fourth digit), therefore the last digit does not have significance in
their course code, and hence will not be used for the definition of the same.
5. Open Electives have a zero in place of the above level numeral and thereby six
digits only.

3
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Course Outline
Semester I

Hours Per Week

S NO Course Title Course Code Credits
L T P

1 Introduction to Artificial Intelligence CAI501C 3 1 0 4

2 Machine Learning CAI502C 3 1 0 4
3 Mathematical Foundations for Machine Learning CAI503C 3 1 0 4

4 Machine Learning Lab CAI504C 0 0 4 2

5 Python Programming Language CAI505C 3 1 0 4

6 Python Programming Language Lab CAI506C 0 0 4 2
7 Elective I/ MOOCs* 3 0 0 3
Total Credits 15 4 4 23

Semester II

Hours Per Week

S NO Course Title Course Code Credits
L T P

1 Optimization Techniques for Machine Learning CAI550C 3 1 0 4

2 Deep Learning CAI551C 3 1 0 4

3 Deep Learning Lab CAI552C 0 0 4 2

4 Applied Statistics And Probability CAI553C 3 1 0 4

5 Exploratory Data Analysis Lab CAI554C 3 0 0 3

6 Elective II/ MOOCs* 3 0 0 3

Total Credits 15 3 2 20

4
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Semester III

Hours Per Week

S NO Course Title Course Code Credits
L T P

1 Deep Reinforcement Learning CAI601C 3 1 0 4

2 Advanced Deep Learning CAI602C 3 1 0 4

3 Research Methodology CAI603C 3 0 0 3

4 Dissertation Work-Phase I CAI604C 0 0 6 3

5 Elective III/ MOOCs* 3 0 0 3

7 Elective IV/ MOOCs* 3 0 0 3

Total Credits 12 2 6 20

Semester IV

Hours Per Week

S NO Course Title Course Code Credits
L P

1 Dissertation Work-Phase II CAI655C 0 28 14

Total Credits 0 14 14

* Subject to proper approval from the departmental monitoring committee (DMC), students will
have the option to earn credits from the MOOCs platforms (SWAYAM/NPTEL) in addition to
choosing courses from the proposed elective baskets. The MOOCs courses are detached from
the semester structure, eliminating the necessity to complete them within the same semester
they're introduced.

5
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

List of Electives for the First Year (Semester I)

Hours Per Week
S No Course Title Course Code Credits
L T P
1 Soft Computing Techniques CAI501E 3 0 0 3

2 Big Data Analytics CAI502E 3 0 0 3

3 Intelligent Information Retrieval CAI503E 3 0 0 3

4 Pattern Recognition CAI504E 3 0 0 3

5 Advanced Algorithms and Analysis CAI505E 3 0 0 3

List of Electives for the First Year (Semester II)

6 Machine Learning for Signal Processing CAI550E 3 0 0 3

7 Electronic Design Automation CAI551E 3 0 0 3

8 Computer Vision CAI552E 3 0 0 3

9 Fuzzy Logic and its Applications CAI553E 3 0 0 3

10 Bio-Inspired Computing CAI554E 3 0 0 3

List of Electives for the Second Year (Semester III)

11 Statistical Modelling for Computer Sciences CAI601E 3 0 0 3

12 Data Engineering CAI602E 3 0 0 3

13 Cognitive Systems CAI603E 3 0 0 3

14 Digital Imaging Techniques and Analysis CAI604E 3 0 0 3

15 Quantum Artificial Intelligence CAI605E 3 0 0 3

16 Natural Language Computing CAI606E 3 0 0 3

17 MLOps CAI607E 3 0 0 3

18 Federated Learning CAI608E 3 0 0 3

6
M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Detailed Syllabus
Core Courses

7
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Introduction to Artificial Intelligence

Course Outcomes
Semester I
Course Code ✔ Acquire foundational knowledge of Artificial Intelligence.
CAI501C ✔ Develop proficiency in various AI problem-solving techniques
✔ Acquire advanced knowledge in AI representation methods
4 Credits
✔ Understand and apply ethical principles in AI design and deployment
L T P
3 1 0

Course Content
UNIT I
AI vs. Human Intelligence, Definition of Artificial Intelligence, Narrow AI vs. General AI, Subfields
of AI, History and Evolution of AI, Applications and limitations of current AI systems, Ethical and
societal challenges of AI
(6 hours)
UNIT II
Intelligent agents: Agents and Environments, the concept of rationality, the nature of
environments, structure of agents, problem solving agents, problem formulation.
Search: informed search , uninformed search , Local search, adversarial search, Constraint
Satisfaction Problems
(8 hours)
UNIT III
Knowledge representation, Logical representation including predicate and first-order logic,
Limitations of logical representation, Semantic networks: structure and usage, Frames: attributes,
slots, values, and inheritance, Scripts for event sequences, Rule-based deduction systems:
forward and backward chaining, resolution, Conceptual graphs: basics and construction,
Reasoning under uncertainty: review of probability, Bayesian networks, Hidden Markov Models
(HMM), Challenges in knowledge representation.
(10 hours)
UNIT IV
Expert systems:- Introduction, basic concepts, structure of expert systems, the human element in
expert systems how expert systems works, problem areas addressed by expert systems, expert
systems success factors, types of expert systems, expert systems and the internet interacts web,
knowledge engineering, scope of knowledge, difficulties, in knowledge acquisition methods of
knowledge acquisition, selecting an appropriate knowledge acquisition method, societal impacts

8
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

reasoning in artificial intelligence, inference with rules, model based reasoning, case based
reasoning, explanation & meta knowledge inference with uncertainty representing uncertainty
(8 hours)
UNIT V
Introduction to AI ethics, key ethical principles in AI: fairness, accountability, transparency,
integrity, sustainability, control, democracy, interoperability, privacy concerns and data protection
in AI systems, bias and discrimination in algorithmic decision-making, AI in surveillance and its
implications for human rights, governance frameworks for AI (national and international
perspectives), role of AI ethics in autonomous and semi-autonomous systems, regulatory and
legal challenges in AI deployment, ethical AI design and implementation practices, case studies
of ethical dilemmas and resolutions in AI applications.
(10 hours)

Text Books

1. Russell, S. J., & Norvig, P. (2010). Artificial Intelligence: A Modern Approach (3rd ed.).
Chapters on Knowledge Representation.

Reference Books

1. Ertel, W. (2018). Introduction to artificial intelligence. Springer.

2. Coeckelbergh, M. (2020). AI ethics. The MIT press essential knowledge series.
3. Wooldridge, M. (2021). A brief history of artificial intelligence: what it is, where we are, and
where we are going. Flatiron Books.

Web References

1. SWAYAM NPTEL 2. MIT OpenCourseWare

An Introduction to Artificial Artificial Intelligence
Intelligence By Patrick Henry Winston
By Prof. Mausam

9
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Machine Learning
Course Outcomes
Semester I
Course Code ✔ Acquire foundational knowledge in machine learning principles.
CAI502C ✔ Develop skills in decision tree algorithms, instance-based learning,
and feature selection techniques.
4 Credits
✔ Master probabilistic models, logistic regression and support vector
L T P machines.
3 1 0 ✔ Implement and analyze artificial neural networks, focusing on
perceptrons and multilayer architectures.
✔ Employ ensemble methods and clustering techniques such as
bagging, boosting, random forests, and k-means.

Course Content
UNIT I
Introduction to Machine Learning, Definition of learning systems, Goals and applications of
machine learning, Different types of learning paradigms, Hypothesis space and inductive bias,
Aspects of developing a learning system: training data, concept representation, function
approximation, Basic Machine learning pipeline – training, testing and validation. Evaluation
Metrics – accuracy, precision, recall, ROC curve. Cross validation, Overfitting, Under fitting,
bias-variance tradeoff, Regularization Theory.
Linear regression, Simple and Multiple Linear regression, Polynomial regression, evaluating
regression model.
(10 hours)
UNIT II
Discriminative Learning, Logistic regression, Multi-class Regression, SoftMax Regression,
Maximum margin classifiers and Support Vector Machines, Hard and soft margin, Higher
dimensional space and Kernel trick. Decision trees – Concept of pure and impure nodes and the
measure of impurity using entropy and Gini index.
(8 hours)
UNIT III
Generative Learning: Bayes Classifier and Naïve Bayes Classifier. Maximum Likelihood Estimation,
Maximum a Posteriori Estimation. Non-Parameterized Density Estimation: Parzen window and
KNN Density estimation.
Clustering: Agglomerative clustering, K-means, Gaussian Mixture models and Expectation
Maximization algorithm.
(8 hours)

10
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT IV
Sequence Modelling and Ensemble Methods: Hidden Markov Models, Conditional Random
Fields. Bagging & boosting and its impact on bias and variance C5.0 boosting, Random forest,
AdaBoost, Gradient Boosting Machines and XGBoost
(6 hours)
UNIT V
Computational learning theory, Introduction to Dimensionality Reduction, The Curse of
Dimensionality, Principal Component Analysis (PCA), Singular Value Decomposition (SVD), Linear
Discriminant Analysis (LDA), Kernel PCA, t-Distributed Stochastic Neighbor Embedding (t-SNE).
(8 hours)

Text Books

1. Machine Learning. Tom Mitchell. McGraw- Hill, 2010.

2. Aurelien Geron, Hands-On Machine Learning with Scikit-Learn and TensorFlow, Oreilly,
March 2017.

Reference Books

1. Alpaydin, Ethem. Introduction to machine learning. MIT press, 2020

2. Kevin P. Murphy, “Machine Learning: A Probabilistic Perspective”, MIT Press
3. Christopher Bishop, “Pattern Recognition and Machine Learning” Springer, 2007.

Web References

1. Coursera DeepLearning.AI 2. Google for Developers

Machine Learning Specialization Google Machine Learning Education
By Andrew Ng, Geoff Ladwig, By Google
Aarti Bagul

11
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Mathematical Foundations for Machine Learning

Semester I Course Outcomes

Course Code ✔ Understand fundamental concepts of linear algebra.

CAI503C ✔ Master dimensionality reduction techniques.
4 Credits ✔ Acquire advanced skills in calculus and optimization.
L T P ✔ Understand basic probability concepts and statistical measures.
3 1 0

Course Content
UNIT I
Systems of equations, Linear Dependence and Independence, Vector Spaces and Subspaces,
Norm of a Vector, Operations with Vectors (Sum, Difference, Scalar Multiplication), Dot Product,
Matrix Operations (Multiplication and Inverse), Determinants, Eigenvalues and Eigenvectors,
Special Matrices and Their Properties, Least Squares and Minimum Norm Solutions, Matrix
Decomposition Algorithms.
(10 hours)
UNIT II
Derivatives: Basic concepts of calculus, Derivative of common functions, Meaning of e and the
derivative of ex, Derivative of log x, Existence of derivatives, Properties of derivative, Partial
derivatives, gradient, directional derivatives, convex function and its properties,chain rule of
differentiation.
(6 hours)
UNIT III
Unconstrained and Constrained optimization, Numerical optimization techniques for constrained
and unconstrained optimization, Optimization using slope method, Optimization using gradient
descent, Adapting Newton’s method for optimization, Penalty function method, Lagrange
Multiplier Method.
(8 hours)
UNIT IV
Basic concepts of probability, conditional probability, independence, Bayes theorem, measures of
central tendency (mean, median, mode), measures of dispersion (variance, standard deviation),
quantiles, box-plots, populations and samples
(8 hours)

12
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT V
Random variables, discrete distributions, continuous distributions, joint distributions, marginal and
conditional distributions, multivariate normal distribution, Central Limit Theorem, unbiased versus
biased estimates, maximum likelihood estimation, confidence intervals, hypothesis testing.
(8 hours)

Text Books

1. M.P. Diesenroth, A. Aldo Faisal, Cheng Soon Ong, Mathematics for Machine Learning,
Cambridge University Press, 2020.

2. W. Cheney, Analysis for Applied Mathematics. New York: Springer Science + Business
Medias, 2001.

Reference Books

1. W. Cheney, Analysis for Applied Mathematics. New York: Springer Science + Business
Medias, 2001.
2. S. Axler, Linear Algebra Done Right (Third Edition). Springer International Publishing, 2015.
3. J. Nocedal and S.J. Wright, Numerical Optimization. New York: Springer Science +
Business Media, 2006.
4. J.S. Rosenthal, A First Look at Rigorous Probability Theory (Second Edition). Singapore:
World Scientific Publishing, 2006.

Web References

1. SWAYAM NPTEL 2. Coursera DeepLearning.AI

Essential Mathematics for Machine Mathematics for Machine Learning and Data
Learning Science Specialization
By Prof. Sanjeev Kumar, Prof. S. K. By Luis Serrano
Gupta | IIT Roorkee

13
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Machine Learning Lab

Course Outcomes
Semester I
Course Code ✔ Develop expertise in data preprocessing techniques essential for
CAI504C cleaning and preparing datasets for machine learning models.
✔ Master feature engineering to enhance model performance by
2 Credits
creating, selecting, and transforming features effectively.
L T P ✔ Apply machine learning algorithms to accurately classify and predict
0 0 4 outcomes across diverse datasets.
✔ Evaluate and fine-tune machine learning models using appropriate
metrics.

LIST OF PRACTICAL EXPERIMENTS

1. Iris Species Classification
Objective: Identify the species of iris plants (setosa, versicolor, or
virginica) based on the measurements of their petals and sepals.
Dataset: The Iris dataset includes 150 records of iris plants, with
four features: the lengths and the widths of the sepals and petals.

2. Titanic Survival Prediction

Objective: Predict whether a passenger survived the Titanic
disaster based on features such as age, gender, and ticket class.
Dataset: The Titanic dataset includes passenger data like name,
age, gender, ticket class, and survival status.

3. Boston Housing Price Prediction

Objective: Predict the median value of homes in various Boston
districts using features like crime rate, property tax rate, and
pupil-teacher ratio.
Dataset: The Boston Housing dataset contains information on
various housing attributes along with the median value of homes
in various areas of Boston.

14
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

4. Diabetes Progression Prediction

Objective: Predict the progression of diabetes in patients one year
after baseline using various diagnostic measurements.
Dataset: The dataset includes measurements such as body mass
index, blood sugar levels, and age, collected from a study on
diabetes progression.

5. Spam Email Detection

Objective: Classify emails as spam or not spam by analyzing their
text content.
Dataset: The dataset typically consists of a collection of email texts
labeled as 'spam' or 'not spam'.

6. MNIST Handwritten Digit Classification

Objective: Recognize handwritten digits (0-9).
Dataset: The MNIST dataset consists of 70,000 grayscale images,
each 28x28 pixels, of handwritten digits. It's divided into a training
set of 60,000 images and a test set of 10,000 images.

7. Movie Recommendation System

Objective: Recommend movies to users based on their past
viewing history and preferences.
Dataset: Datasets for this task often include user ratings for
various movies, which can be used to learn preference profiles for
individual users.

8. Stock Prices Prediction

Objective: Predict future stock prices based on historical price and
volume data.
Dataset: Historical stock prices data, which includes daily opening,
closing, highest, and lowest prices, and volume of stocks traded.

9. Sentiment Analysis of Text

Objective: Determine the sentiment (positive, negative, neutral) of
a piece of text, such as a tweet or a product review.
Dataset: This involves datasets containing text data with
corresponding sentiment labels.

15
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

10. Fashion-MNIST Classification

Objective: Classify images of clothing items into 10 categories
(e.g., T-shirts, trousers, shoes).
Dataset: Fashion-MNIST is a dataset comprising 70,000 grayscale
images of 10 fashion categories, with each image being 28x28
pixels.

Text Books

1. Aurelien Geron, Hands-On Machine Learning with Scikit-Learn and TensorFlow, Oreilly,
March 2017.

Reference Books

1. Dr. M Gopal, Applied Machine Learning, 1st Edition, McGraw-Hill,2018

2. Aurelien Geron, Hands-On Machine Learning with Scikit-Learn and TensorFlow, Oreilly,
March 2017.

Web References

1. freeCodeCamp 2. codebasics Youtube

Machine Learning for Everybody Machine Learning Tutorial Python
By Kylie Ying By Dhaval Patel

16
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Python Programming Language

Course Outcomes
Semester I
Course Code ✔ Develop a strong foundation in the basic Python programming
CAI505C language.
✔ Master object-oriented programming concepts within Python.
4 Credits
✔ Gain expertise in data analysis using NumPy and Pandas.
L T P ✔ Acquire skills to visualize data effectively using Matplotlib and
3 1 0 Seaborn.
✔ Gain hands-on experience with advanced topics such as design
patterns and Python web frameworks.

Course Content
UNIT I
Introduction to Development Environments: Familiarization with Jupyter Notebooks and Python
IDEs like PyCharm and Visual Studio Code. Python Basics: Syntax, Variables, Data Types, and
Control Structures. Functions and Modules: Defining Functions, Scope, and Importing Modules.
Data Structures: Lists, Tuples, Sets, Dictionaries, and Comprehensions. File Handling: Techniques
for Reading from and Writing to Files.
(8 hours)
UNIT II
Introduction to OOP Concepts, Classes and Objects, Attributes and Methods, Constructors
(__init__ method), Encapsulation, Inheritance, Polymorphism, Special Methods (like __str__ and
__repr__), Class and Static Methods, Property Decorators, Composition vs Inheritance

(8 hours)
UNIT III
NumPy: Understanding ndarrays, Data Types and Attributes, Array Creation and Properties,
Indexing and Slicing, Array Mathematics (Addition, Subtraction, Scalar Multiplication, Division),
Aggregation and Statistical Functions, Array Manipulation (Reshape, Concatenate, Split).
Pandas: Series and DataFrames, Data Importing and Exporting, Data Cleaning and Preparation,
Data Manipulation (Indexing, Selection, Filtering), Working with Missing Data, GroupBy
Operations, Merging and Joining DataFrames, Reshaping and Pivoting.
(8 hours)
UNIT IV
Basic Plotting with Matplotlib (Line Graphs, Bar Charts, Histograms, Scatter Plots), Customizing
Plots (Colors, Labels, Legends), Advanced Plotting Techniques (Subplots, 3D Plots, Interactive

17
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Visualizations), Introduction to Seaborn, Seaborn’s Built-In Datasets, Statistical Plotting with

Seaborn (Distribution Plots, Categorical Plots, Pair Plots, Heatmaps)
(8 hours)
UNIT V
Iterators and Generators, Decorators, Context Managers, Regular Expressions, Testing and
Debugging (using unittest), Virtual Environments, Introduction to Python Web Frameworks (
Flask), Design Patterns
(8 hours)
Text Books

1. Downey, A. (2012). Think python. " O'Reilly Media, Inc.".

Reference Books
1. Shaw, Z. A. (2024). Learn Python The Hard Way. Addison-Wesley Professional.
2. Sweigart, A. (2016). Invent your own computer games with python.
3. Barry, P. (2016). Head first Python: A brain-friendly guide. " O'Reilly Media, Inc.".
4. Matthes, E. (2023). Python crash course: A hands-on, project-based introduction to
programming.

Web References

1. Code with Mosh 2. Harvard University

Python Tutorial - Python Full Course CS50
for Beginners Introduction to Computer Science
By Mosh Hamedani By David J. Malan

18
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Python Programming Language Lab

Course Outcomes
Semester I
Course Code ✔ Master Python basics, data manipulation with Pandas, and numerical
CAI506C operations with NumPy.
✔ Learn data visualization using Matplotlib and Seaborn.
2 Credits
✔ Understand and apply key machine learning algorithms using
L T P Scikit-learn.
0 0 4 ✔ Gain skills in model evaluation, performance metrics and
hyperparameter tuning.
✔ Get introduced to deep learning with TensorFlow and Keras,
culminating in a comprehensive final project.

LIST OF PRACTICAL EXPERIMENTS

1. Basic Syntax and Script Writing
Experiment: Write a simple Python script that takes user input, processes it, and outputs a
result, such as a script that calculates the area of a circle given its radius.
2. Data Types and Variables
Experiment: Create variables of different data types (integer, float, string, list, tuple,
dictionary) and perform basic operations on them, like adding numbers or
concatenating/joining strings.
3. Control Flow
Experiment: Write programs that use if, elif, and else statements to make decisions, and
use for and while loops to iterate over sequences or repeat actions until a condition is
met.
4. Functions and Modules
Experiment: Define functions to perform specific tasks. Also, learn to use Python modules
by importing and using functions from the standard library.
5. File Handling
Experiment: Read from and write to files in Python. Create a script that reads a text file
and counts the frequency of each word in the file.
6. Error Handling and Exceptions
Experiment: Write a program that handles different types of exceptions, such as handling
division by zero or handling file operations when a file does not exist.
7. Classes and Object-Oriented Programming
Experiment: Create a class representing a simple concept, such as a Book with attributes
like title and author, and methods to display book info.

19
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

8. List Comprehensions and Generators

Experiment: Use list comprehensions to create lists in a single line of code. For example,
create a list of squares of the first 10 natural numbers. Also, experiment with generators to
generate an infinite sequence.
9. Decorators and Higher-Order Functions
Experiment: Write decorators to modify existing functions, such as a decorator that logs
function calls or measures the execution time of functions.
10. Regular Expressions
Experiment: Use regular expressions to perform complex string matching and extraction,
such as extracting all email addresses from a large text.
11. Web Scraping
Experiment: Use libraries like BeautifulSoup or Scrapy to scrape data from web pages.
12. Web Development
Experiment: Build a simple web application using a framework like Flask.
13. Database Interaction
Experiment: Connect to a SQL database using sqlite3 or another database library and
perform CRUD operations.
14. Data Analysis and Visualization
Experiment: Use pandas and matplotlib to analyze a dataset and create visualizations like
histograms and scatter plots.
15. Asynchronous Programming
Experiment: Write asynchronous code using asyncio to perform multiple tasks
concurrently.
16. Script Packaging and Distribution
Experiment: Package a Python script and distribute it as a package/installable module.

Text Books
1. Downey, A. (2012). Think python. " O'Reilly Media, Inc.".

Reference Books
1. Shaw, Z. A. (2024). Learn Python The Hard Way. Addison-Wesley Professional.
2. Sweigart, A. (2016). Invent your own computer games with python.
3. Barry, P. (2016). Head first Python: A brain-friendly guide. " O'Reilly Media, Inc.".
4. Matthes, E. (2023). Python crash course: A hands-on, project-based introduction to
programming.

20
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Optimization Techniques for Machine Learning

Course Outcomes
Semester II
Course Code ✔ To understand the theory of optimization methods and algorithms
CAI550C developed for solving various types of optimization problems
✔ To develop and promote research interest in applying optimization
4 Credits
techniques in problems of engineering and technology
L T P ✔ To apply the mathematical results and numerical techniques of
3 1 0 optimization theory to concrete engineering problems.

Course Content
UNIT I
Foundations: convex sets, functions, Taylor series and local function approximation, Numerical
methods for computing derivatives, automatic differentiation, Numerical computation of
eigenvalues / singular values, conjugates, subdifferentials, weak and strong duality, Introduction
to optimisation problems, Optimality conditions, Nonconvex optimality & stationarity, Tractable
nonconvex problems,
(10 hours)
UNIT II
Unconstrained optimization: Local search as a general optimization paradigm, Random local
search and the curse of dimensionality, Steplength rules and intuition, Formal principles of
random local search
(8 hours)
UNIT III
First order methods: Gradient descent - normalized and unnormalized versions, Step size
selection and convergence, Conservative theoretical guarantees and linesearch, Formal
principles of gradient descent, Steepest descent - norm generalization, Batch / Stochastic
gradient descent, Variations on normalized stochastic gradient descent
(8 hours)
UNIT IV
Second order methods: Newton’s method basics - normalized and unnormalized versions,
Newton’s method, stepsize, and backtracking linesearch, Honest adjustments for non-convex
functions, Scaling issues with Newton’s method, Newton and secant method as zero-finding
algorithms, Quasi-Newton Methods, BFGS and L-BFGS algorithms
(6 hours)

21
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT V
Methods of constrained optimization: Projected / proximal gradient algorithms, Barrier and
penalty methods, Interior point methods, Primal-dual methods, Duality and Lagrangian
(8 hours)
Text Books

1. Nowozin, S., Wright, S. J., & Sra, S. (Eds.). (2011). Optimization for Machine Learning. MIT
Press.

Reference Books
1. Brownlee, J. (2021). Optimization for machine learning. Machine Learning Mastery.
2. Aggarwal, C. C., Aggarwal, L. F., & Lagerstrom-Fife. (2020). Linear algebra and
optimization for machine learning (Vol. 156). Springer International Publishing.

22
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Deep Learning
Course Outcomes
Semester II
Course Code ✔ Able to introduce deep learning and application of modern neural
CAI551C networks.
✔ Understand the deep learning algorithms extract layered
4 Credits
representations of data.
L T P ✔ Utilize Deep Neural Networks for Image Analysis.
3 1 0 ✔ Explore the Foundations and Applications of Deep Learning, the
Driving Force Behind Recent AI Advancements.

Course Content
UNIT I
History of Deep Learning, McCulloch Pitts Neuron, Perceptrons, Perceptron Learning Algorithm,
Multilayer Perceptrons (MLPs), Representation
Power of MLPs, Sigmoid Neurons, Gradient Descent, Feedforward Neural Networks (FFNs),
Representation Power of FFNs, Backpropagation
(8 hours)
UNIT II
Optimization algorithms and activation functions: Gradient Descent (GD), Momentum based GD,
stochastic GD, mini-batch GD, Adagrad, RMSProp, Adam.
Initialization techniques: Xavier and He initialization.
Regularization: Bias Variance Tradeoff, L2 regularization, Early stopping, Dataset augmentation,
Parameter sharing and tying, Injecting noise at input, Dropout, Batch Normalization,
(8 hours)
UNIT III
Convolutional Neural Networks (CNN): Convolution operation, filters, Padding and Stride, Sparse
Connectivity and Weight Sharing, Max Pooling and NonLinearities.
Transfer Learning and pretrained CNN architectures: AlexNet, ZFNet, VGGNet, GoogleNet,
ResNet. Batch Normalization, Dropout.
(8 hours)

23
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT IV
Basic Concepts in Object Detection: Bounding box and annotation techniques, Non-maximum
suppression (NMS), R-CNN and its evolution (Fast R-CNN, Faster R-CNN), You Only Look Once
(YOLO) series, Single Shot MultiBox Detector (SSD)
Semantic Segmentation, U-Net and its variants image segmentation, SegNet and its architecture,
Instance and Panoptic Segmentation, Mask R-CNN for instance segmentation, Metrics for
performance evaluation (mAP for detection, IoU for segmentation)
(8 hours)
UNIT V
Recurrent Neural Networks (RNN): Sequence Learning problems, Intuition behind RNN, sequence
classification, sequence labeling, Model, Loss function, Learning algorithm, Evaluation. Vanishing
and Exploding gradient. LSTMs and GRUs, Encoder Decoder models, Attention mechanism,
Graph Neural Networks (GNNs).
(8 hours)

Text Books

1. Bengio, Yoshua, Ian J. Goodfellow, and Aaron Courville. "Deep learning." An MIT Press
book in preparation. (2015) .

Reference Books

1. Bengio, Yoshua. "Learning deep architectures for AI." Foundations and trends in Machine
Learning 2.1 (2009): 1127.
2. Géron, A. (2022). Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow. "
O'Reilly Media, Inc.".
3. Trask, A. W. (2019). Grokking deep learning. Simon and Schuster.

Web References

1. NPTEL 2. MIT 6.S191

Deep Learning Introduction to Deep Learning
By Mitesh M. Khapra By MIT

24
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Deep Learning Lab

Course Outcomes
Semester II
Course Code ✔ To Build The Foundation Of Deep Learning.
CAI552C ✔ To understand How to Build The Neural Network.
✔ To enable students to develop successful machine learning
2 Credits
concepts.
L T P
0 0 4

LIST OF PRACTICAL EXPERIMENTS

1. Implementing Gradient Descent Algorithm from Scratch
Objective: Understand and implement the gradient descent optimization algorithm to
minimize a simple cost function.

2. Data Preprocessing
Objective: Load, reshape, normalize, and preprocess data for a neural network model.
This includes converting labels to one-hot encoding

3. Building and Training Neural Networks using Tesorflow

Objective: Build, train, validate, and infer with a neural network using Keras, and learn to
save and reload the model

4. Building and Training Neural Networks using PyTorch

Objective: Build, train, validate, and infer with a neural network using PyTorch, and learn to
save and reload the model

5. Binary Classification of Images

Objective: Create a CNN that can differentiate between cat and dog images.
Dataset: Cats vs. Dogs Dataset (commonly found on Kaggle)

6. Multiclass Classification of Images

Objective: Build a simple convolutional neural network (CNN) to classify images
Datasets: MNIST, Fashion-MNIST and CIFAR-10

7. Implementing Data Augmentation

25
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Objective: Apply data augmentation techniques to enhance the training dataset for a
neural network, improving model robustness and helping prevent overfitting.

8. Transfer Learning for Image Classification

Objective: Utilize a pre-trained model (like VGG16, ResNet, or MobileNet) as a feature
extractor and fine-tune it to classify a new set of images.
Dataset: Use the Oxford 102 Flowers dataset for flower classification or the Stanford Cars
dataset for car classification.

9. Sentiment Analysis
Objective: Train a neural network to classify movie reviews from the IMDB dataset as
positive or negative.
Dataset: IMDB Movie Reviews

10. Stock Prices Prediction

Objective: Build a model using RNNs to predict future stock prices based on historical
price data.
Dataset: Any stock price historical data

11. Language Detection

Objective: Train a neural network to detect the language of a given text snippet.
Dataset: WiLI-2018, a benchmark dataset for language identification

12. Generative Adversarial Network (GAN)

Objective: Generate digits by training a GAN on Identify the Digits (MNIST) dataset
Dataset: Identify the Digits (MNIST)

13. Graph Neural Network (GNN)

Objective: Implement and explore basic Graph Neural Network (GNN) architectures to
solve problems related to molecular data
Dataset: MoleculeNet (Tox-21)

Text Books

1. Chollet, François. "Deep Learning with Python," Manning Publications, 2017.

Reference Books

1. Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville, MIT Press.
2. The Elements of Statistical Learning by T. Hastie, R. Tibshirani, and J. Friedman, Springer.
3. Probabilistic Graphical Models. Koller, and N. Friedman, MIT Press.

26
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Applied Statistics and Probability

Course Outcomes
Semester II
Course Code ✔ Understand the fundamental principles of statistics and their
CAI553C application in engineering, including the design and analysis of
experiments and observational studies.
4 Credits
✔ Master measures Of Central tendency and dispersion to describe
L T P data distributions effectively
3 1 0 ✔ Gain expertise in key probability distribution and their applications
in engineering problems.
✔ Develop the ability to conduct and interpret hypothesis testing,
including the use of t-tests, F-tests, and Chi-Square tests for data
analysis.

Course Content
UNIT I
Basics of Statistics: The Role of Statistics in Engineering, Basic Principles, Retrospective Study,
Observational study, Designed Experiments, Observing Processes over time, Mechanistic and
Empirical Models, Probability and Probability Models
(8 hours)
UNIT II
Measures of central tendency: mean, median, and mode; Measures of dispersion, Range, Quartile
Deviation, Mean Deviation, Standard Deviation, Coefficient of variance, Skewness, Kurtosis.
(8 hours)
UNIT III
Probability Distribution: Sample Spaces and Events, Interpretations and Axioms of Probability,
Addition Rules, Conditional Probability, Total Probability. Random Variables, Concept of Random
Variable, Bernoulli Distribution, Binomial Distribution, Poisson Distribution, Normal Distribution.
(8 hours)
UNIT IV
Correlation and Regression: Concept and types, Karl Pearson Method, Rank Spearman Method,
Least Square Method, Discrete Random Variables and Probability Distributions. Continuous
Random Variables and Probability Distributions. Joint Probability Distributions.
(8 hours)

27
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT V
Testing of Hypothesis: Testing of Hypothesis, Null and alternative hypothesis, level of
significance, one-tailed and two-tailed tests, tests for large samples (tests for single mean,
difference of means, single proportion, difference of proportions), t-test, F- test, Chi-Square Test.
(8 hours)
Text Books
1. Douglas C. Montgomery and George C. Runger, Applied Statistics and Probability for
Engineers, John Wiley & Sons. 2016

Reference Books

1. Blake I., An Introduction to Applied Probability, John Wiley & Sons.

2. Yagolam A. M. and Yagolam I. M., Probability and Information, Hindustan Publishing
Corporation, Delhi, 1983.
3. Statistics for Management - Richard I Levin, David S Rubin - Prentice Hall India –6thEdn,
ISBN-81-203-0893-X.

28
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Exploratory Data Analysis Lab

Course Outcomes
Semester II
Course Code ✔ Gain proficiency in Python-based data visualization (Matplotlib,
CAI554C Pandas, Seaborn) and introductory Tableau Desktop skills.
✔ Master data aggregation and manipulation techniques, including
1 Credits
pivot tables and the split-apply-combine strategy.
L T P ✔ Acquire foundational skills in creating charts, performing data
0 0 2 filtering and sorting in Tableau
✔ Learn to analyze relationships between quantitative factors using
scatterplots, and regression diagnostics.

LIST OF PRACTICAL EXPERIMENTS

1. Basic Plotting:
Create line plots and scatter plots using Matplotlib.
Customize plots with labels, legends, and annotations.
2. Advanced Graph Types:
Generate histograms, box plots, and bar charts to visualize distributions and comparisons.
Explore the use of subplots to display multiple plots in one figure.
3. Interactive Visualizations:
Implement interactive elements such as buttons and sliders to modify plots in real time.
4. Statistical Data Visualization:
Use Seaborn to create visually appealing statistical plots like violin plots, swarm plots, and
pair plots.
Understand how to integrate Seaborn with Matplotlib for enhanced customization.
5. Facet Grids and Categorical Data Visualization:
Employ FacetGrid to create a grid of plots based on a dataset’s features.
Visualize categorical data using bar plots, box plots, and count plots.
6. Basic Interactive Charts:
Create interactive charts like line charts, scatter plots, and area charts using Plotly.
Explore Plotly's capabilities to modify aspects of the chart interactively.
7. Complex Interactive Visualizations:
Develop complex visualizations such as 3D plots and geographic data maps.
Use Dash by Plotly to create web-based interactive dashboards.
8. Integrating Plotly with Web Applications:
Learn how to embed Plotly visualizations into web applications.

29
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Explore interactive features that allow users to manipulate data or change visualizations
through web interfaces
9. Summary Statistics and Correlation Analysis
Generate summary statistics tables and heatmaps of correlations between variables.
10. Principal Component Analysis (PCA)
Apply PCA to reduce dimensions and visualize the dataset in two or three dimensions.
11. Missing Data Analysis
Visualize and quantify missing data in datasets.
12. Outlier Detection
Use scatter plots and box plots to identify outliers in datasets.
13. Time Series Analysis
Explore autocorrelation and trends over time with line plots and rolling averages.
14. Distribution Analysis
Use kernel density plots and cumulative distribution functions (CDFs) to analyze the
underlying distributions of variables.
Create Q-Q plots to assess the normality of the distributions.
15. Bivariate Analysis
Generate pair grids to visualize relationships between each pair of variables.
Use bubble charts to represent three variables on two dimensions, with the size of the
bubble adding an additional dimension
16. Multivariate Analysis
Conduct cluster analysis to identify natural groupings in the data.
Visualize high-dimensional data using t-distributed Stochastic Neighbor Embedding
(t-SNE).
17. Text Data Exploration
Utilize word clouds to visualize the most frequent terms in textual data.
Conduct sentiment analysis to gauge the sentiment expressed in text data and visualize
the results.
18. Calculate and visualize PDFs and CDFs to understand the probability distributions of
different variables within the dataset

Text Books
1. Tamara Munzner, Visualization Analysis & Design, CRC Press, 2014.
Reference Books

1. Scott Murray, Interactive Data Visualization for the Web, (2e), O’Reilly, 2017.

30
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

2. Wingston Chang, R graphics cookbook, O'Reilly. (2013).

3. Andy Field, Jeremy Miles, and Zoe Field, Discovering Statistics Using R, SAGE
Publications Ltd. 2012.

31
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Deep Reinforcement Learning

Course Outcomes
Semester III
Course Code ✔ Understand the core principles of reinforcement learning
CAI601C ✔ Understand the model decision-making process using Markov
Decision making processes
4 Credits
✔ Understand and apply exploration and exploitation strategies in
L T P reinforcement learning
3 1 0 ✔ Acquire Skills in Monte Carlo methods for practical policy
evaluation and control
✔ Understand function approximation in the context of Reinforcement
Learning

Course Content
UNIT I
Overview of reinforcement learning, Difference between supervised, unsupervised, and
reinforcement learning, Components of reinforcement learning (agent, environment, state, action,
reward, transaction function, Discount, episode), Bandit problem.
(5 hours)
UNIT II
Understanding Markov Decision Processes (MDPs), Policy, State value function, Action value
function, Bellman equation, Dynamic Programming—Policy iteration, Policy Improvement, Value
iteration, and Limitations of dynamic programming.
(9 hours)
UNIT III
Exploration and exploitation strategies- Random, Greedy, Epsilon-Greedy, Softmax, UCB.
Monte Carlo Methods: First-Visit Monte Carlo, Every-Visit Monte Carlo, Monte Carlo simulation for
policy evaluation.
(9 hours)
UNIT IV
Temporal Difference Learning (TD), n-step TD, Q-learning, SARSA, bootstrapping.
(8 hours)
UNIT V
Function Approximation, linear function approximation, Deep Q-networks (DQN), Double Deep
Q-networks (DDQN), Dueling Q-networks (Dueling DQN), Policy gradient methods, and Actor
Critic methods.
(9 hours)

32
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books

1. “Reinforcement Learning: An Introduction” by Richard S. Sutton and Andrew G. Barto.

Reference Books

1. “Grokking Deep Reinforcement Learning” by Miguel Morales, 2020

2. Alexander Zai , Brandon Brown, Deep Reinforcement Learning in Action, 2020, 1st Edition,
Manning Publications.
3. Mohit Sewak, Deep Reinforcement Learning: Frontiers of Artificial Intelligence, 2019,
Springer.
4. Sugiyama, Masashi, Statistical reinforcement learning: modern machine learning, 2015,
Chapman and Hall

33
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Advanced Deep Learning

Course Outcomes
Semester III
Course Code ✔ Acquire knowledge to design advanced deep learning models to
CAI602C address novel challenges in practical applications
✔ Understand and apply the concept of Meta Learning
4 Credits
✔ Gain skills in implementing Transformer architectures and LLMs
L T P ✔ Develop the ability to design and train various types of GANs
3 1 0 ✔ Learn to implement and innovate with Graph Neural Networks

Course Content
UNIT I
Meta-learning: Concepts of Few-Shot Learning, One-Shot Learning, and Zero-Shot Learning,
Meta-learning Algorithms (Model-Agnostic Meta-Learning (MAML), Prototypical Networks,
Matching Networks), Meta-Reinforcement Learning, Applications of Meta-learning in Real-World
Scenarios, Neural Architecture Search, Hyperparameter Optimization, Meta-Learning for Fast
Adaptation of Deep Networks, Continuous Adaptation via Meta-Learning in Nonstationary and
Competitive Environments, Learning to Generalize: Meta-Learning for Domain Generalization
(8 hours)
UNIT II
Attention Mechanisms and Transformers: Introduction to Attention Mechanisms, Types of
Attention (Scaled Dot-Product Attention, Multi-Head Attention), The Architecture of Transformers,
Self-Attention Mechanism, Positional Encoding, Encoder-Decoder Structure in Transformers,
Application of Transformers in Natural Language Processing (NLP), Fine-Tuning Pretrained Vision
Transformers, Large Language Models (LLMs), Applications of LLMs
(8 hours)
UNIT III
Generative Adversarial Networks: Introduction to GANs, Understanding Generators and
Discriminators, Theoretical Foundations of GANs, Types of GANs (DCGAN, CGAN, InfoGAN,
CycleGAN), Training Stability and Techniques, Loss Functions in GANs, Evaluation Metrics for
GANs, StyleGAN, BigGAN, RadialGAN, Relativistic discriminator
(8 hours)
UNIT IV
Learning on graphs: Graph Representation Techniques, Node Embedding Techniques, Graph
Neural Networks (GNNs), Spectral Methods for Graph Analysis, Spatial Methods for Graph
Learning, Applications of Graph Learning (Social Network Analysis, Recommendation Systems,

34
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Fraud Detection), Key Architectures in GNNs Graph Convolutional Networks, Semi-Supervised

Classification with Graph Convolutional Networks, Gated Graph Sequence Neural Networks,
Graph Attention Networks, GraphSAGE, Inductive Representation Learning on Large Graphs,
Hierarchical Graph Representation Learning with Differentiable Pooling, GraphRNN: Generating
Realistic Graphs with Deep Auto-regressive Models, Learning Graphical State Transitions
(8 hours)
UNIT V
Introduction to Explainability and Interpretability, Techniques for Model Interpretability, Feature
importance methods, permutation importance, Partial dependence plots (PDPs), Local
interpretable model-agnostic explanations (LIME), SHapley Additive exPlanations (SHAP),
Interpretability in Computer vision (saliency maps and Grad-CAM), Interpretability in Natural
language processing (attention mechanisms), Quantitative Testing with Concept Activation
Vectors (TCAV), Improving Simple Models with Confidence Profiles
(8 hours)
Text Books
1. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press.

Reference Books

1. Vasilev, I. (2019). Advanced Deep Learning with Python: Design and implement advanced
next-generation AI solutions using TensorFlow and PyTorch. Packt Publishing Ltd.
2. Trask, A. W. (2019). Grokking deep learning. Simon and Schuster.
3. Aurelien Geron, Hands-On Machine Learning with Scikit-Learn and TensorFlow, Oreilly,
March 2017.

35
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Research Methodology
Course Outcomes
Semester III
Course Code ✔ Develop a foundational understanding of scientific research
CAI603C ✔ Master scientific writing and effective communication
✔ Acquire technical skills in scientific documentation
3 Credits
✔ Understand the use of databases and research metrics in scientific
L T P research
3 0 0

Course Content
UNIT I
Science and Scientific Research: Knowledge and the epistemology of knowledge, deductive and
inductive inference, a brief history of scientific ideas, important thinkers and scientific
advancements, principles of effective research, self-development in research, the creative
process, roles of the problem-solver and problem-creator, identifying and solving new problems,
literature survey, developing a research plan, writing research proposals.
(8 hours)
UNIT II
Scientific Writing and Communication: Steps in writing a research report, layout of the research
report, writing references and bibliography, importance and planning of effective presentations,
how to write good scientific papers, models of the paper writing process, benefits of targeting
reputable journals, peer review process, responding to reviewer comments, reviewing papers,
identifying publication misconduct, dealing with complaints and appeals with publishers and
journals.
(8 hours)
UNIT III
Technical and Scientific Documentation Using LaTeX: Introduction to LaTeX, short history and
main attractions of LaTeX, automatic styling according to journal requirements, cross-referencing,
writing complex mathematical expressions, typical LaTeX input files, the edit/format/preview
process, embedding references, bibliography management using BibTeX, creating presentations
using Beamer, introduction to Overleaf, practical sessions on LaTeX usage.
(8 hours)
UNIT IV
Databases: Overview of indexing databases, citation databases like Web of Science and Scopus.

36
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Research Metrics : Understanding metrics including Journal Impact Factor, SNIP, R-Index,
CiteScore, h-index, g-index, i10-index, altmetrics.
(8 hours)
Text Books
1. Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed
methods approaches. Sage Publications.

Reference Books

1. Booth, W. C., Colomb, G. G., & Williams, J. M. (2009). The craft of research. University of
Chicago press.
2. Schimel, J. (2012). Writing Science: How to Write Papers That Get Cited and Proposals
That Get Funded. Oxford University Press.

37
M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Detailed Syllabus
Electives

38
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Soft Computing Techniques

Course Code Course Outcomes
CAI501E
✔ Know about the concepts of Fuzzy logic, crisp logic, fuzzy relation,
Credits
fuzzy implication rule
L T P ✔ Know about optimization theory, genetic computing and
3 0 0 evolutionary computing.
✔ Know about the concepts of the neural network, Perceptron,
implementation and training

Course Content
UNIT I
Introduction of soft computing: What is Soft Computing, soft computing vs. hard computing, soft
computing paradigms, and applications of soft computing. Basics of Machine Learning. Dealing
with Imprecision and Uncertainty- Probabilistic Reasoning- Bayesian network, Pearl’s Scheme for
Evidential Reasoning, Dempster-Shafer Theory for Uncertainty Management, Certainty Factor
based Reasoning
(7 hours)
UNIT II
Neural Networks: Basics of Neural Networks- Neural Network Structure and Function of a single
neuron: Biological neuron, artificial neuron, definition of ANN, Taxonomy of neural net,
characteristics and applications of ANN, McCulloch Pitt model, different activation functions,
Supervised Learning algorithms- Perceptron (Single Layer, Multi-layer), Linear separability,
ADALINE, MADALINE, RBF networks , Widrow Hoff, learning rule, Delta learning rule, Back
Propagation algorithm, Un-Supervised Learning algorithms- Hebbian Learning, Winner take all.
(9 hours)
UNIT III
Fuzzy Logic: Fuzzy set theory, Fuzzy set versus crisp set, Crisp relation & fuzzy relations, Fuzzy
systems: crisp logic, fuzzy logic, introduction & features of membership functions, Fuzzy rule base
system: fuzzy propositions, formation, decomposition & aggregation of fuzzy rules, fuzzy
reasoning, fuzzy inference systems, Mamdani Fuzzy Models – Sugeno Fuzzy Models, Adaptive
Neuro-Fuzzy Inference Systems Architecture.
(6 hours)

39
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT IV
Optimization: Derivative-based Optimization – Descent Methods – The Method of Steepest
Descent – Classical Newton’s Method, Simulated Annealing, Random Search, Downhill Simplex
Search Derivative-free Optimization- Genetic algorithm Fundamentals, basic concepts, working
principle, encoding, fitness function, reproduction, Genetic modeling: Inheritance operator, cross
over, mutation operator, Generational Cycle, Convergence of GA, Applications & advances in GA,
Differences & similarities between GA & other traditional methods.
(8 hours)

Text Books
1. S, Rajasekaran & G.A. Vijayalakshmi Pai, “Neural Networks, Fuzzy Logic & Genetic
Algorithms, Synthesis & Applications”, PHI Publication.
2. Jyh-Shing Roger Jang, Chuen-Tsai Sun, EijiMizutani, “Neuro-Fuzzy and Soft Computing”,
Prentice-Hall of India. 1996

Reference Books

1. SAndries P Engelbrecht, Computational Intelligence: An Introduction, Wiley

Publications.2007
2. S.N. Sivanandam & S.N. Deepa, “Principles of Soft Computing”, Wiley Publications.2018

40
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Big Data Analytics

Course Code Course Outcomes
CAI502E
✔ To work with big data platforms and explore big data analytic
3 Credits
techniques for business applications.
L T P ✔ To design efficient algorithms for mining the data from large
3 0 0 volumes.
✔ To Analyze the HADOOP and Map Reduce technologies associated
with big data analytics.
✔ To explore Big Data Applications Using Pig and Hive.

Course Content
UNIT I
Introduction to Big Data Platform – Challenges of Conventional Systems - Intelligent data analysis
– Nature of Data - Analytic Processes and Tools - Analysis vs Reporting.
(5 hours)
UNIT II
Introduction To Streams Concepts – Stream Data Model and Architecture - Stream Computing -
Sampling Data in a Stream – Filtering Streams – Counting Distinct Elements in a Stream –
Estimating Moments – Counting Oneness in a Window – Decaying Window - Real time Analytics
Platform (RTAP) Applications - Case Studies - Real Time Sentiment Analysis- Stock Market
Predictions.
(10 hours)
UNIT III
History of Hadoop- the Hadoop Distributed File System – Components of Hadoop Analyzing the
Data with Hadoop- Scaling Out- Hadoop Streaming- Design of HDFS-Java interfaces to HDFS
Basics- Developing a Map Reduce Application-How Map Reduce Works-Anatomy of a Map
Reduce Job run-Failures-Job Scheduling-Shuffle and Sort – Task execution - Map Reduce Types
and Formats- Map Reduce Features- Hadoop environment.
(10 hours)
UNIT IV
Applications on Big Data Using Pig and Hive – Data processing operators in Pig – Hive services –
HiveQL – Querying Data in Hive - fundamentals of HBase and ZooKeeper - IBM InfoSphere
BigInsights and Streams.
(5 hours)

41
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books
1. Tom White “Hadoop: The Definitive Guide” Third Edition, O’reilly Media, 2012.
2. Seema Acharya, Subhasini Chellappan,Big Data Analytics, Wiley 2015.

Reference Books
1. Michael Berthold, David J. Hand, “Intelligent Data Analysis”, Springer, 2007.
2. Chris Eaton, Dirk DeRoos, Tom Deutsch, George Lapis, Paul Zikopoulos, “Understanding
Big Data: Analytics for Enterprise Class Hadoop and Streaming Data”, McGrawHill
Publishing, 2012.
3. Anand Rajaraman and Jeffrey David Ullman, “Mining of Massive Datasets”, CUP, 2012.
4. Bill Franks, “Taming the Big Data Tidal Wave: Finding Opportunities in Huge Data Streams
with Advanced Analytics”, John Wiley& sons, 2012.
5. Glenn J. Myatt, “Making Sense of Data”, John Wiley & Sons, 2007.
6. Pete Warden, “Big Data Glossary”, O’Reilly, 2011.

42
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Intelligent Information retrieval

Course Code Course Outcomes
CAI503E
✔ Describe the genesis and variety of information retrieval situations.
3 Credits
✔ Construct a variety of information retrieval models and techniques.
L T P ✔ Execute methods and principles of information retrieval system.
3 0 0 ✔ Develop Methods for implementing information retrieval systems.
✔ Evaluate the emerging information retrieval practices in library
services and on the Web.

Course Content
UNIT I
Fundamentals of IR Systems, Models and Indexing: Overview of IR Systems, Information retrieval
using the Boolean model, the dictionary and postings lists, Tolerant retrieval, Automatic Indexing,
Index construction and compression, Scoring, Vector space model and term weighting.
Document Representation and Analysis: Statistical Characteristics of Text, Regular Expressions,
Text Normalization, Edit Distance, N-Gram Language Models, Naive Bayes and Sentiment
Classification-Logistic Regression for Document Analysis
(10 hours)
UNIT II
Query Processing and Evaluation: Basic Query Processing, Data Structure and File Organization
for IR, Evaluation in information retrieval-Relevance feedback, User Profiles, Collaborative
Filtering and query expansion.
(7 hours)
UNIT III
Retrieval Models: Similarity Measures and Ranking, Boolean Matching, Vector Space Models,
Probabilistic Models, XML Retrieval, Language models for information retrieval.
(6 hours)
UNIT IV
Web Search Analysis: Web search basics, web characteristics, index size and estimation, near
duplicates and shingling, web crawling, distributing indexes, connectivity servers, link analysis,
web as a graph, PageRank, Hubs and authoritative pages, summarization, question answering

(7 hours)

43
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books
1. C. D. Manning, P. Raghavan, and H. Schutze, Introduction to Information Retrieval,
Cambridge University Press (2008).

Reference Books

1. Ricardo Baezce Yates, Berthier Ribeiro-Neto, Modern Information Retrieval: The Concepts
and Technology behind Search (2ndEd, 2010).
2. Mikhail Klassen, Matthew A. Russell, Mining the Social Web, O'Reilly Media, Inc., 3rd
Edition (2019)

44
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Advanced Algorithms and Analysis

Course Outcomes
Semester I
Course Code ✔ Understand and apply key algorithmic strategies with basic
CAI504C programming skills.
✔ Differentiate between computational complexity classes and tackle
4 Credits
NP-Complete problems.
L T P ✔ Employ probabilistic and amortized analysis to evaluate algorithm
3 1 0 performance.
✔ Design approximation algorithms for NP-hard challenges, including
basic parallel computing techniques,

Course Content
UNIT I
Defining Key Terms: Algorithm complexity, Greedy method, Dynamic Programming, Backtracking,
Branch-and-bound Techniques; Examples for understanding above techniques; Memory model,
linked lists and basic programming skills.
(8 hours)
UNIT II
Overview - Class P - Class NP - NP Hardness - NP Completeness - Cook Levine Theorem -
Important NP Complete Problems. Heuristic and Randomized algorithms.
(8 hours)
UNIT III
Use of probabilistic inequalities in analysis, Amortized Analysis - Aggregate Method - Accounting
Method - Potential Method, competitive analysis, applications using examples.
(8 hours)
UNIT IV
Point location, Convex hulls and Voronoi diagrams, Arrangements, graph connectivity, Network
Flow and Matching: Flow Algorithms - Maximum Flow – Cuts - Maximum Bipartite Matching -
Graph partitioning via multi-commodity flow, Karger's Min Cut Algorithm, String matching and
document processing algorithms.
(8 hours)
UNIT V
Approximation algorithms for known NP hard problems - Analysis of Approximation Algorithms
Use of Linear programming and primal dual; local search heuristics; Parallel algorithms: Basic
techniques for sorting, searching, merging, list ranking in PRAMs and Interconnection.
(8 hours)

45
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books
1. Michael T Goodric and Roberto Tamassia, “Algorithm Design: Foundations, Analysis and
Internet Examples”, John Wiley and Sons, 2002.
2. Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest and Clifford Stein, “Introduction
to Algorithms”, Third Edition, The MIT Press, 2009.

Reference Books
1. Allan Borodin and Ran El-Yaniv: Online Computation and Competitive Analysis, Cambridge
University Press, 2005.
2. Sanjoy Dasgupta, Christos Papadimitriou and Umesh Vazirani, “Algorithms”, Tata McGraw-
Hill, 2009.
3. RK Ahuja, TL Magnanti and JB Orlin, “Network flows: Theory, Algorithms, and
Applications”, Prentice Hall Englewood Cliffs, NJ 1993.

Web References

1. freeCodeCamp 2. Harvard University

Data Structures and Algorithms with CS 224
Visualizations – Full Course (Java) Advanced Algorithms
By Dinesh Varyani By Harvard

46
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Pattern Recognition
Course Code Course Outcomes
CAI504E
✔ Recognize the characteristics of machine learning strategies.
3 Credits
✔ Apply various supervised learning methods to appropriate
L T P problems.
3 0 0 ✔ Identify and integrate more than one technique to enhance the
performance of learning.
✔ Create probabilistic and unsupervised learning models handling
unknown patterns.
✔ Analyze the co-occurrences of the data to find frequent patterns

Course Content
UNIT I
Pattern Classifier: Overview of pattern recognition - Discriminant functions - Supervised learning -
Parametric estimation - Maximum likelihood estimation - Bayesian parameter estimation-
Perceptron algorithm - LMSE algorithm - Problems with Bayes approach - Pattern classification by
distance functions - Minimum distance pattern classifier.
(11 hours)
UNIT II
Unsupervised Classification: Clustering for unsupervised learning and classification - Clustering
concept - C-means algorithm – Hierarchical clustering procedures - Graph theoretic approach to
pattern clustering - Validity of clustering solutions.
(9 hours)
UNIT III
Structural Pattern Recognition Elements of formal grammars - String generation as pattern
description - Recognition of syntactic description - Parsing -Stochastic grammars and applications
- Graph based structural representation.
(5 hours)
UNIT IV
Feature Extraction and Selection: Entropy minimization - Karhunen - Loeve transformation -
Feature selection through functions approximation - Binary feature selection.
(5 hours)

47
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books
1. Robert Schalkoff, Pattern Recognition: Statistical Structural and Neural Approaches Wiley–
India, 2009
2. Theodoridis, S. and K. Koutroumbas, “Pattern Recognition”, Fourth Edition, San Diego, CA:
Academic Press, 2009.
Reference Books
1. Robert J.Schalkoff, Pattern Recognition: Statistical, Structural and Neural Approaches,
John Wiley & Sons Inc., New York, 1992.
2. Tou and Gonzales, Pattern Recognition Principles, Wesley Publication Company, London,
1974.
3. Duda R.O., and Hart.P.E., Pattern Classification and Scene Analysis, Wiley, New York, 1973.
4. Morton Nadier and Eric Smith P., Pattern Recognition Engineering, John Wiley & Sons,
New York, 1993.
5. Christopher M Bishop, Pattern Recognition and Machine Learning. Springer. 2011.

48
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Machine Learning for Signal processing

Course Code Course Outcomes
CAI550E
✔ Apply linear algebra and probability theory to solve problems in
3 Credits
audio signal processing.
L T P ✔ Understand and apply machine learning algorithms to tasks in
3 0 0 music information retrieval and speech recognition.
✔ Develop and evaluate models for various audio processing
applications like genre classification, event detection, and speaker
diarization.

Course Content
UNIT I
Linear Algebra Refresher: Matrices, vectors, eigenvalues, and eigenvectors. Programming Basics:
Python for data manipulation, bash scripting for task automation. Digital Signal Processing for
Audio: Time-frequency analysis, filtering, and feature extraction.
(8 hours)
UNIT II
Probability Theory Refresher: Bayesian inference, random variables, distributions. Machine
Learning Basics: Supervised and unsupervised learning paradigms, evaluation metrics.
(6 hours)

UNIT III
Music Information Retrieval: Feature extraction from music, similarity measures, retrieval systems.
Classification and Tagging: Genre classification, mood detection, auto-tagging systems.
(8 hours)
UNIT IV
Speech Recognition: Feature extraction from speech, hidden Markov models, deep learning
approaches. Other Audio Processing Applications: Acoustic event detection, speaker diarization,
query by humming, melody estimation.
(8 hours)
Text Books
1. “Automatic Speech Recognition: A Deep Learning Approach”, D. Yu and L. Deng, Springer,
2016.
2. “Pattern Recognition and Machine Learning”, C.M. Bishop, 2nd Edition, Springer, 2011.

49
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Reference Books
1. “Deep Learning”, I. Goodfellow, Y, Bengio, A. Courville, MIT Press, 2016.
2. “An Introduction to Audio Content Analysis”, A. Lerch, Wiley-IEEE Press, 2012.
3. “Speech and audio signal processing: processing and perception of speech and music”,
B. Gold, N. Morgan, D. Ellis, Wiley, 2011
4. “Signal Processing Methods for Music Transcription”, A. Klapuri and M. Davy, Springer,
2007.

50
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Electronic Design Automation

Course Code Course Outcomes
CAI551E
✔ Understand the Foundations of EDA.
3 Credits
✔ Apply General-Purpose Methods for Combinatorial Optimization.
L T P ✔ Model and Simulate Electronic Designs
3 0 0 ✔ Conduct High-Level Synthesis and Understand FPGA Physical
Design Automation

Course Content
UNIT I
PRELIMINARIES: Introduction to design methodologies, Design Automation tools, Algorithmic
Graph theory, Computational complexity, tractable and intractable problems.
(4 hours)
UNIT II
GENERAL PURPOSE METHODS FOR COMBINATORIAL OPTIMIZATION: Backtracking branch and
bound, Dynamic programming, Integer linear programming, Local search, simulated annealing,
tabu search, Genetic Algorithms.
Layout compaction, Placement, Floor planning and Routing problems, Concepts and algorithms.
(8 hours)
UNIT III
MODELING AND SIMULATION: Gate level modeling and simulation, Switch level modeling and
simulations.
LOGIC SYNTHESIS AND VERIFICATION: Basic issues and terminology, Binary decision diagrams,
two–level logic synthesis.
(8 hours)
UNIT IV
HIGH–LEVEL SYNTHESIS: Hardware models, Internal representation of the input algorithm,
Allocation assignment and scheduling, some Scheduling Algorithms, some suspects of
Assignment problem, High level transformations.
PHYSICAL DESIGN AUTOMATION OF FPGA TECHNOLOGIES, Physical design cycle for FPGAs,
Partitioning and Routing for segmented and staggered models.
(10 hours)

51
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books

1. Algorithms for VLSI design, S.H Gerez, WILEY student edition, John Wiley & Sons (Asia)
Pvt. Ltd.,1999.

Reference Books

1. Algorithms for VLSI physical design automation, 3rd edition, Naveed sherwani, springer
international edition, 2005
2. Computer aided logical design with emphasis on VLSI-Hill & Peterson, Wiley.
3. M.J.S.Smith, “Application Specific Integrated Circuits”,Pearson, 2008.

52
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Computer Vision
Course Code Course Outcomes
CAI552E
✔ Understand foundational principles of computer vision
3 Credits
✔ Apply techniques in image segmentation and recognition,
L T P extracting meaningful patterns from visual data through shape and
3 0 0 texture analysis.
✔ Utilize computer vision methods for analyzing and managing
real-world video data, employing generative models for innovative
visual content creation.

Course Content
UNIT I
Introduction to Computer Vision, The Four Rs of Computer Vision, Low-level vs High-level
processing, Two View Geometry, Binocular Stereopsis, Camera and Epipolar Geometry Image
Formation: Planar Scenes and Homography, Depth Estimation, Robust Correspondence Image
Representation: Fourier Transform, Feature Detection, Edge Detection, Local Binary Patterns
Recognition: Pyramid Matching, Part-based recognition models.
(8 hours)
UNIT II
Evolution of CNN Architectures: AlexNet, MobileNet, InceptionNets, ResNets, DenseNets
Advanced Neural Networks: SIFT & Single Object Recognition, Dense Neural Networks Image
Enhancements: Image Quality Enhancement, Image Restoration, Super-resolution Machine
Learning Techniques: Unsupervised Learning, Reinforcement Learning in Vision, Salient
Detection.
(8 hours)
UNIT III
Image Segmentation: Supervised Segmentation, Agglomerative Clustering, UNet, FCN
Recognition and Description: Dense Descriptors, Optical Flow & Tracking, Visual Matching:
Bag-of-words Shape and Texture Analysis: Shape from Texture, Color, Motion and Edges, Face
Detection.
(6 hours)

53
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT IV
Video Analytics: Crowd Analysis, Video Surveillance, Traffic Monitoring Deep Generative Models:
GANs, VAEs, Zero-shot, One-shot Learning Recognition and Retrieval: Content-based Image
Retrieval, Instance Recognition Anomaly Detection and Recognition: Anomalous Action
Recognition, Post Estimation.
(6 hours)

Text Books

1. Computer Vision: Algorithms and Applications, 2nd ed.Richard Szeliski, The University of
Washington,2022

Reference Books

1. Computer Vision: A Modern Approach (Second Edition) by David Forsyth and Jean Ponce
2. D. A. Forsyth and J. Ponce, Computer Vision, A Modern Approach, Pearson Education,
2003.

54
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Fuzzy Logic and Its Applications

Course Code Course Outcomes
CAI553E
✔ Understand the fundamentals of Fuzzy logic and its applications
3 Credits
✔ Apply the concepts of fuzzy logic to solve real-world problems
L T P ✔ Design fuzzy systems for various engineering applications
3 0 0 ✔ Analyze the performance of fuzzy systems

Course Content
UNIT I
Different faces of imprecision - inexactness, Ambiguity, Undecidability, Fuzzyness and certainty,
Fuzzy sets and crisp sets, Probability and fuzzy logic, Fuzzy control and knowledge based
systems
(7 hours)
UNIT II
Impressive concepts, Fuzzyness and imprecision, Properties of fuzzy sets, Fuzzy representation,
Conventional set operations, Intersection of Fuzzy sets, Union of fuzzy sets, the complement of
fuzzy sets
(7 hours)
UNIT III
Linguistic variables, Fuzzy propositions, Fuzzy compositional rules of inference-the-Min-Max rules
implications and fuzzy additive rules of implication, Methods of decompositions and
defuzzification -composite moments, composite maximum average of maximum values and
center of maximums
(8 hours)
UNIT IV
Direct and Indirect methods with single and multiple experts, Construction from sample data -
Least square method, adaptive fuzzy controllers - membership function tuning using gradient
descent
(8 hours)

55
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books

1. Zimmermann H.J., ‘Fuzzy Set Theory - and its Applications’, Springer, 4th Ed., 2007
2. Timothy J. Ross, ‘Fuzzy Logic with Engineering Applications’, Wiley Publications, 4th Ed.,
2016

Reference Books

1. John Yen, Reza Langari, ‘Fuzzy Logic, Intelligence, Control & Information’, Pearson
Education Inc., India, 2007
2. Zdenko Kovacic, Stjepan Bogdan, ‘Fuzzy Controller Design Theory and Applications’, CRC
Press, 1st Ed., 2006
3. Riza C. Berkaan, Sheldon L. Trubatch, ‘Fuzzy Systems Design Principles – Building Fuzzy
IF THEN Rule Based’, IEEE Press, 1997
4. George J Klir and Bo Yuan, ‘Fuzzy Sets and Fuzzy Logic: Theory and A: Theory and
Applications’, Pearson, 2015
5. M. Mitchell, ‘Introduction to Genetic Algorithms’, Indian Reprint, MIT press Cambridge, 2nd
Ed, 2014

56
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Bio-Inspired Computing
Course Outcomes
Semester II
Course Code ✔ Understand basic concepts of evolutionary computing.
CAI554C ✔ Understand basic features of neural and immune systems and be
able to build the neural network
4 Credits
✔ Explain how complex and functional high-level phenomena emerge
L T P from low-level interactions.
3 1 0 ✔ Explain the computational processes derived from neural models.
✔ Implement simple bio-inspired algorithms like genetic and Particle
Swarm Optimization

Course Content
UNIT I
INTRODUCTION TO EVOLUTIONARY ALGORITHM: Evolutionary algorithm, components of
evolutionary algorithm representation (definition of individuals), Evaluation function (Fitness
function), Population, parent selection Mechanism, Variation Operators, Survivor Selection
Mechanism (Replacement), Initialization, Termination Condition, evolutionary algorithm case study
Cellular systems, cellular automata, modeling with cellular systems, other cellular systems,
computation with cellular systems, artificial life: analysis and synthesis of cellular systems.
(6 hours)
UNIT II
NEURAL SYSTEMS: Biological nervous systems, artificial neural networks, neuron models,
architecture, signal encoding, synaptic plasticity, unsupervised learning, supervised learning,
reinforcement learning, evolution of neural networks, hybrid neural systems, case study.
DEVELOPMENTAL AND IMMUNE SYSTEMS: Rewriting system, synthesis of developmental
system, evolutionary rewriting systems, evolutionary developmental programs, biological immune
systems, lessons for artificial immune systems, algorithms and applications, shape space,
negative selection algorithm, clonal selection algorithm.
(7 hours)
UNIT III
BEHAVIORAL SYSTEMS: Behavior is cognitive science, behavior in AI, behavior-based robotics,
biological inspiration for robots, robots as biological models, robot learning, evolution of
behavioral systems, learning in behavioral systems, co-evolution of body and control, towards
self-reproduction, simulation and reality.
(6 hours)

57
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

UNIT IV
GENETIC ALGORITHMS: Representation of Individuals, Mutation, Recombination, Population
Models, Parent Selection, Survivor Selection, Example Application: Solving a Job Shop
Scheduling Problem
HYBRIDIZATION WITH OTHER TECHNIQUES: MEMETIC ALGORITHMS: Introduction to Local
Search, Lamarckianism and the Baldwin Effect, Structure of a Memetic Algorithm, Heuristic or
Intelligent Initialization, Hybridization within Variation Operators: Intelligent Crossover and
Mutation, Local Search Acting on the output from Variation Operators, Hybridization During the
Genotype to Phenotype Mapping, Design Issues for Memetic Algorithms.
(10 hours)
UNIT V
COLLECTIVE SYSTEMS: Biological self-organization, Particle Swarm Optimization (PSO), ant
colony optimization (ACO), swarm robotics, co-evolutionary dynamics, artificial evolution of
competing systems, artificial evolution of cooperation, case study.
(8 hours)
Text Books

1. D. Floreano and C. Mattiussi, “Bio-Inspired Artificial Intelligence”, MIT Press, 2008.

2. Tao Song, Pan Zheng, Mou Ling Dennis Wong, Xun Wang, “Bio-Inspired Computing
Models and Algorithms”, ISBN: 978-981-3143-19-7, world scientific, 2019 F.
3. Neumann and C. Witt, “Bioinspired Computation in combinatorial optimization: Algorithms
and their computational complexity”, Springer, 2010.

Reference Books
1. D. Floreano and C. Mattiussi, "Bio-Inspired Artificial Intelligence", MIT Press, 2008.
2. Tao Song, Pan Zheng, Mou Ling Dennis Wong, Xun Wang, “Bio-Inspired Computing
Models and Algorithms”, ISBN: 978-981-3143-19-7, world scientific, 2019 F.
3. Neumann and C. Witt, “Bioinspired Computation in combinatorial optimization: Algorithms
and their computational complexity”, Springer, 2010.
4. D. E. Goldberg, “Genetic algorithms in search, optimization, and machine learning”,
Addison- Wesley, 1989.
5. Simon O. Haykin, “Neural Networks and Learning Machines”, Third Edition, Prentice Hall,
2008

58
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Statistical Modelling for Computer Sciences

Course Code Course Outcomes
CAI601E
✔ To implement statistical analysis techniques for solving practical
3 Credits
problems.
L T P ✔ To perform statistical analysis on a variety of data.
3 0 0 ✔ To perform appropriate statistical tests and visualize the outcome

Course Content
UNIT I
Probability Theory: Sample Spaces- Events - Axioms – Counting - Conditional Probability and
Bayes’ Theorem – The Binomial Theorem – Random variable and distributions: Mean and
Variance of a Random Variable-Binomial-Poisson-Exponential and Normal distributions. Curve
Fitting and Principles of Least Squares- Regression and correlation.
(8 hours)
UNIT II
Sampling Distributions & Descriptive Statistics: The Central Limit Theorem, distributions of the
sample mean and the sample variance for a normal population, Sampling distributions
(Chi-Square, t, F, z). Test of Hypothesis- Testing for Attributes – Mean of Normal Population –
One-tailed and two-tailed tests, F-test and Chi-Square test - - Analysis of variance ANOVA – One
way and two way classifications.
(8 hours)
UNIT III
Tabular data- Power and the computation of sample size- Advanced data handling- Multiple
regression- Linear models- Logistic regression- Rates and Poisson regression- Nonlinear curve
fitting.
(6 hours)
UNIT IV
Density Estimation- Recursive Partitioning- Smoothers and Generalized Additive Models-
Survivals Analysis- Analyzing Longitudinal Data- Simultaneous Inference and Multiple
Comparisons- Meta-Analysis- Principal Component Analysis- Multidimensional Scaling- Cluster
Analysis
(8 hours)

59
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books

1. Dalgaard, Peter, “Introductory statistics with R”, Springer Science & Business Media, 2008.
2. Brain S. Everitt, “A Handbook of Statistical Analysis Using R”, Second Edition,LLC, 2014.

Reference Books
1. Richard Cotton, “Learning R”, O’Reilly, 2013.
2. Samir Madhavan, “Mastering Python for Data Science”, Packt, 2015.
3. Sheldon M. Ross,”Introduction to Probability and Statistics for Engineers and Scientists”,
4th edition, Academic Press; 2009.
4. Paul Teetor, “R Cookbook, O’Reilly, 2011.

60
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Data Engineering
Course Code Course Outcomes
CAI602E
✔ Master key data storage and processing technologies
3 Credits
✔ Develop expertise in designing and implementing robust ETL
L T P pipelines
3 0 0 ✔ Gain proficiency in data orchestration

Course Content
UNIT I
Overview of Data Engineering: Definition, importance, and role in business and technology.
Data Systems and Architecture: Introduction to databases, Modern Data Ecosystem, data
warehouses, data lakes, and data marts. Data Engineering vs. Data Science: Distinctions between
the roles and how they complement each other.
(6 hours)
UNIT II
Data Storage and Processing: Types of Data, Understanding Different Types of File Formats
Relational Databases (RDBMS): Principles, design, and SQL.
NoSQL Databases: Types (Key-Value, Document, Columnar, Graph), use cases, and examples.
Foundations of Big Data: Hadoop ecosystem, Apache Spark, HDFS and Hive. Data Modeling:
Concepts of normalization, denormalization, schema design, and data partitioning.
(8 hours)
UNIT III
ETL, ELT, and Data Pipelines, Data Integration and Data Integration Platforms
Data Extraction Techniques: APIs, web scraping, database queries.
Data Transformation: Data cleaning, validation, aggregation, and transformation techniques.
Data Loading: Techniques for efficient data loading into various storage systems.
(8 hours)
UNIT IV
Architecting the Data Platform, Factors for Selecting and Designing Data Stores, Importance of
Data Security, How to Gather and Import Data, Data Wrangling, Tools for Data Wrangling,
Querying and Analyzing Data. Data Orchestration and Monitoring: Workflow Management:
Apache Airflow. Scheduling and Automation: Techniques and tools for automating and scheduling
data pipelines. Monitoring and Logging: Best practices for monitoring data pipelines, logging, and
error handling. (8 hours)

61
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books

1. Reis, J., & Housley, M. (2022). Fundamentals of Data Engineering. " O'Reilly Media, Inc.".

Reference Books

1. Kleppmann, M. (2017). Designing data-intensive applications: The big ideas behind

reliable, scalable, and maintainable systems. " O'Reilly Media, Inc.".
2. Warren, J., & Marz, N. (2015). Big Data: Principles and best practices of scalable realtime
data systems. Simon and Schuster.

62
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Cognitive Systems
Course Code Course Outcomes
CAI603E
✔ Understand the basics Of Cognitive Computing and its differences
3 Credits
from traditional approaches to Computing.
L T P ✔ Plan and use the primary tools associated with cognitive
3 0 0 computing.
✔ Plan and execute a project that leverages Cognitive Computing.

Course Content
UNIT I
Cognitive science and cognitive Computing with AI, Cognitive Computing - Cognitive Psychology
The Architecture of the Mind - The Nature of Cognitive Psychology – Cognitive architecture
Cognitive processes – The Cognitive Modeling Paradigms - Declarative / Logic based
Computational cognitive modeling – connectionist models – Bayesian models. Introduction to
Knowledge-Based AI – Human Cognition on AI – Cognitive Architectures
(8 hours)
UNIT II
Decision making, Fuzzy Cognitive Maps, Learning algorithms: Nonlinear Hebbian Learning – Data
driven NHL - Hybrid learning, Fuzzy Gray cognitive maps, Dynamic Random fuzzy cognitive Maps.
(8 hours)
UNIT III
Machine learning Techniques for cognitive decision making – Hypothesis Generation and Scoring
- Natural Language Processing - Representing Knowledge - Taxonomies and Ontologies - Deep
Learning.
(7 hours)
UNIT IV
Cognitive Systems in health care – Cognitive Assistant for visually impaired – AI for cancer
detection, Predictive Analytics - Text Analytics - Image Analytics -Speech Analytics – IBM Watson
(7 hours)

Text Books

1. Hurwitz, Kaufman, and Bowles, “Cognitive Computing and Big Data Analytics”, Wiley,
Indianapolis.
2. Neil Stillings, Steven E. Weisler, Christopher H. Chase and Mark H. Feinstein, “Cognitive
Science: An Introduction”, MITPress.

63
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Reference Books

1. Peter Fingar, Cognitive Computing: A Brief Guide for Game Changers, PHI Publication,
2015.
2. Gerardus Blokdyk ,Cognitive Computing Complete Self-Assessment Guide, 2018
3. Rob High, Tanmay Bakshi, Cognitive Computing with IBM Watson: Build smart applications
using Artificial Intelligence as a service, IBM Book Series, 2019

64
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Digital Imaging Techniques and Analysis

Course Code Course Outcomes
CAI604E
✔ Ascertain and describe the essentials of image processing
3 Credits
concepts through mathematical interpretation.
L T P ✔ Experiment with various image segmentation and morphological
3 0 0 operations for a meaningful partition of objects.
✔ Design the various basic feature extraction and selection
procedures for various image applications.

Course Content
UNIT I
INTRODUCTION TO IMAGE PROCESSING: Introduction, Digital Image Fundamentals, image
acquisition and display using digital devices - Human visual perception, properties – Image
Formation - Image sampling and quantization-Basic relationship between pixels.
(5 hours)
UNIT II
IMAGE ENHANCEMENT: Image enhancement in the spatial domain: basic gray level
transformation, Histogram Processing- Enhancement using arithmetic/Logic operations, Spatial
filtering: smoothing and sharpening. Image enhancement in the frequency domain: Introduction
to two-dimensional transforms- Discrete Fourier Transform, Discrete Cosine Transform, Haar
Transform, Discrete Wavelet Transform - smoothing frequency domain filtering-sharpening
frequency domain filtering.
(11 hours)
UNIT III
MORPHOLOGICAL IMAGE PROCESSING: Morphological Image Processing: Dilation and Erosion
– Opening and Closing – Hit or Miss Transformation – Thinning – Thickening – Skeleton.
(5 hours)
UNIT IV
IMAGE SEGMENTATION: Image Segmentation: Detection of discontinuities- Object Detection
Methods, Edge Linking and Boundary Detection, Thresholding Methods, Region Oriented
Methods.
FEATURE EXTRACTION: Region of interest (ROI) selection - Feature extraction: Histogram based
features - Intensity Features-Color, Shape Features-Local Binary Patterns (LBP), Texture
descriptors- Grey Level Occurrence Matrix (GLCM).
(9 hours)

65
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Text Books

1. Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing”, Pearson Education,

Fourth Edition, 2018.

Reference Books

1. S. Sridhar, “Digital Image Processing”, Second Edition, Oxford University, 2016.

2. Anil K. Jain “Fundamentals of Digital Image Processing”, PHI, Learning Private Ltd, 2011.
3. Milan Sonka, Vaciav Hlavac, Roger Boyle, “Image Processing Analysis and Vision”, Fourth
Edition, Cengage India, 2017

66
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Quantum Artificial Intelligence

Course Code Course Outcomes
CAI605E
✔ Able to solve machine learning problems using quantum
3 Credits
computations.
L T P ✔ Knows the use of quantum logic for data mining and applications.
3 0 0 ✔ Independently understand and solve multi level problems in
machine learning

Course Content
UNIT I
Introduction: Learning theory and data mining, quantum like classical computers.
(6 hours)
UNIT II
Data driven models, supervised and unsupervised learning, generalization performance,
ensembles, data dependencies and examples.
(8 hours)
UNIT III
Pattern Recognition and Neural Networks, perception, Hopfield Networks, Feedforward networks,
Deep learning, computational complexity.
(8 hours)
UNIT IV
Quantum Pattern Recognition, Quantum Associative Memory, Quantum Perceptron, Quantum
Neural Networks, Physical Realizations.
(8 hours)

Text Books

1. Isaac Chuang, Michael Nielsen, Quantum Computation and Quantum Information, 10th
Anniversary Edition, Cambridge University Press, 2011.
2. Maria Schuld, Ilya Sinayskiy, Francesco Petruccione, An introduction to quantum machine
learning, 2014.

67
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Reference Books

1. P. Wittek, Quantum Machine Learning, Elesevier, Amsterdam 2014.

2. Articles in https://arxiv.org/qu-comp

3. S. Bhattacharyya, I. Pan, A. Mani, S. De, E. Behrman, S. Chakraborti (Eds.), Quantum

Machine Learning, Walter de Gruyter, Berlin, 2020

68
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Natural Language Computing

Course Outcomes

Course Code ✔ To get introduced to language processing technologies for

CAI606E processing text data.
✔ To understand the role of Information Retrieval and Information
3 Credits
Extraction in Text Analytics
L T P ✔ To acquire knowledge on text data analytics using language
3 0 0 models.

Course Content
UNIT I
Natural Language Processing – Linguistic Background – Mathematical Foundations -
Morphological Analysis-Tokenization - Stemming-Lemmatization – Boundary Determination.
(8 hours)
UNIT II
Reading unstructured data - Representing text data - Part of speech tagging – Syntactic
representation -Text similarity - WordNet based similarity- Shallow parsing -Semantic
representation.
(8 hours)
UNIT III
Information retrieval and Information extraction - Named Entity Recognition – Relation
Identification-Template filling.
(6 hours)
UNIT IV
Language model - Probabilistic Models - n-gram language models- Hidden Markov Model- Topic
Modelling - Graph Models -Feature Selection and classifiers -Rule-based Classifiers - Maximum
entropy classifier – Clustering-Word and Phrase-based Clustering.
(8 hours)

Text Books

1. Christopher D. Manning and Hinrich Schutze,“Foundations of Statistical Natural Language

Processing”, MIT Press, 1999.

69
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Reference Books

1. Steven Struhl, “Practical Text Analytics: Interpreting Text and Unstructured Data for
Business Intelligence”, Kogan Page, 2015.
2. Matthew A. Russell, “Mining the Social Web”, O'Reilly Media, 2013.
3. Steven Bird, Ewan Klein and Edward Loper, “Natural Language Processing with Python”,
1st Edition, O'Reilly Media, 2009.

70
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

MLOps
Course Outcomes

Course Code ✔ Understand the significance of MLOps

CAI607E ✔ Learn to deploy machine learning models using different strategies
✔ Gain skills in implementing CI/CD pipelines
3 Credits
✔ Develop skills in containerization and manage ML deployments
L T P
3 0 0

Course Content
UNIT I
Overview of MLOps and its importance, Difference between MLOps, DevOps, and DataOps, Key
principles and best practices, Lifecycle stages: from model development to deployment,
Managing data and model versioning, Experiment tracking and model reproducibility
(8 hours)
UNIT II
Techniques for handling large datasets, Data versioning tools like DVC, Integrating data quality
checks in workflows, Automating model training processes, Scalable model training techniques,
Continuous integration and delivery (CI/CD) for ML models, Evaluation metrics and validation
strategies
(8 hours)
UNIT III
Deployment strategies (online, batch, streaming), Model serving frameworks (TensorFlow Serving,
TorchServe), Canary deployments and A/B testing, Monitoring model performance in production,
Detecting and handling model drift, Strategies for model retraining and updating
(8 hours)
UNIT IV
Containerization with Docker, Kubernetes for scalable ML deployments, Managing compute
resources, Implementing feature stores for machine learning, ML workflow orchestration with
tools like Kubeflow, Airflow
(8 hours)
Text Books
1. Treveil, M., Omont, N., Stenac, C., Lefevre, K., Phan, D., Zentici, J., ... & Heidmann, L.
(2020). Introducing MLOps. O'Reilly Media.

71
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Reference Books

1. Ameisen, E. (2020). Building Machine Learning Powered Applications: Going from Idea to
Product. " O'Reilly Media, Inc.".
2. Gift, N., & Deza, A. (2021). Practical MLOps. " O'Reilly Media, Inc.".
3. Huyen, C. (2022). Designing machine learning systems. " O'Reilly Media, Inc.".

72
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Federated Learning
Course Outcomes

Course Code ✔ Knowledge of the basic concepts, architecture, and applications of

CAI608E Federated Learning
✔ Analyze horizontal federated learning
3 Credits
✔ Understand the significance of Federated Learning for Vision,
L T P Language, and Recommendation
3 0 0

Course Content
UNIT I
Basics of Machine Learning :Overview of Machine Learning principles, Distributed Machine,
Learning: Introduction to distributed systems, need for distributed learning, Data parallelism vs
model parallelism, Fundamentals of Federated Learning: Definition and goals of Federated
Learning, Key challenges and opportunities, Types of Federated Learning: Cross-device vs.
Cross-silo, Vertical vs. Horizontal, Architecture and Communication Protocols: Client-server
architecture, Communication protocols and efficiency, Aggregation strategies (e.g., Federated
Averaging) Privacy and Security: Differential privacy, Secure multi-party computation,
Privacy-preserving techniques in Federated Learning
(8 hours)
UNIT II
Federated Learning Optimization Algorithms: Federated Averaging (FedAvg), Variants of
Federated Averaging, Optimization challenges and solutions, Statistical and system heterogenity,
Personalization and Adaptation: Personalization Strategies in Federated Learning, Federated
meta-learning, Transfer learning in Federated contexts, Robustness and Fault Tolerance: Handling
non-IID data, Dealing with unreliable clients and communication failures, Robust aggregation
methods, Federated Learning with Limited Resources: Resource-efficient algorithms, Model
compression and quantization, Trade-offs between accuracy and efficiency
(8 hours)
UNIT III
Frameworks and Tools: Overview of Federated Learning frameworks (e.g., Flower, TensorFlow
Federated, PySyft), Data Preprocessing and Management: Data partitioning and preparation for
Federated Learning (IID and non-IID data), Handling heterogeneous data sources, Data privacy
and anonymization techniques, Model Training and Evaluation: Training machine learning models

73
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

in a Federated setup, Evaluation metrics for Federated Learning, Monitoring and debugging
Federated Learning processes
(8 hours)
UNIT IV
Case study: Google Gboard, Advanced Topics: Federated transfer learning, Federated analytics,
Federated LLM’s, poisoning in FL, Decentralised Federated Learning
(8 hours)
Text Books
1. Yang, Q., Liu, Y., Cheng, Y., Kang, Y., Chen, T., & Yu, H. (2019). Federated Learning. Morgan
& Claypool Publishers.

Reference Books

1. Jin, Y., Zhu, H., Xu, J., & Chen, Y. (2023). Federated Learning: Fundamentals and
Advances. Springer Nature Singapore.

74
 M. Tech (AI) Syllabus, Centre for AI, IUST-Kashmir

Centre for Artificial Intelligence

75