Multidisciplinary

Minor-9
Generative AI
 Page 2 of 17

Preface

Generative AI represents a dynamic field at the forefront of technological

advancement, where algorithms are imbued with the capacity to imagine, create,
and innovate. In this course, we aim to provide you with a comprehensive
understanding of the principles, techniques, and applications of Generative AI,
while offering you the opportunity to delve into specialized topics and hands-on
projects tailored to your interests.
The Computer Science & Engineering department offers a Multidisciplinary Minor
in Generative AI to thestudents of Civil, Computer Engineering, Electronics &
Telecommunication, and Mechanical Engineering starting from the third semester
itself. It spans over five semesters, culminating with a capstone project, adding up
to a total of 14 credits.
The course starts with the understanding of Artificial intelligence and machine
learning and getting familiar with the concepts of neural networks, deep learning,
which provides the necessary foundation for diving deeper into Generative AI.
Students will also learn various models in Generative AI and how these models
generate new data points by learning the underlying distribution of the training
data. Also they willget acquainted with various tools, techniques and applications
of Generative AI. Students will gain hands-on experience by implementing
generative models using popular deep learning frameworks like TensorFlow or
PyTorch along with different architectures, loss functions, and hyperparameters to
understand their impact on the quality of generated outputs. Explore creative
applications of GenerativeAI beyond generating realistic images or text. This could
include music generation, style transfer, or even generating entire stories or
artworks. Students will also learn ethical implications of Generative AI, such as
deepfakes and data privacy concerns. Also they will understand the societal impact
of these technologies and how they can be used responsibly.
Objectives
● To provide fundamental knowledge of AI and ML
● To prepare the students with demanding industry skills
● To provide an opportunity to develop expertise in AI tools & technologies.
● To apply theoretical understanding to hands-on interdisciplinary
projects, solving problems usingGenerative AI models

Outcomes
● Familiar with basic AI and ML pipeline
● Apply 25+ AI tools to various business models.
● Generate innovative ideas, contents & outputs for industry applications.

Department of Multidisciplinary Studies MDM-9

 Page 3 of 17

CURRICULUM STRUCTURE

Generative AI
Credit Scheme Teaching Scheme
Course Evaluation Scheme and Marks
(Hours/Week)
Code
T
L P T T L P T FA
Course Name o
o
t
t FA1 FA2 SA TW PR OR Total
a
a
l
l

Semester III

BCS23MD Introduction to
01 2 - - 2 2 - - 10 10 30 - - - 50
Generative AI

Semester IV

BCS24MD Generative AI
02 2 - - 2 2 - - 10 10 30 - - - 50
Models

Semester V

BCS25MD Generative AI
03 Tools & 3 - - 3 3 - - 20 20 60 - - - 100
Techniques
BCS25MD Generative AI
04 Tools & - 1 - 1 - 2 - - - - - - 50 50
Techniques
Laboratory

Semester VI

BCS26MD Generative AI
05 2 - - 2 2 - - 10 10 30 - - - 50
Applications

Semester VII

BCS27MD Capstone
06 - 4 - 4 - 8 - - - - 100 - 50 150
Project
Total
9 5 - 14 9 10 - 450

L-Lecture, P-Practical, T-Tutorial, FA-Formative Assessment, SA-Summative Assessment, TW-Term

Work, OR-Oral, PR-Practical

Department of Multidisciplinary Studies MDM-9

 Page 4 of 17

MDM: B. Tech.: Civil/Comp/CompR/E&Tc/Mech/IT Semester: III

Generative AI

Course: Introduction to Generative AI Code: BCS23MD01

Teaching Scheme (Hrs./Week) Evaluation Scheme and Marks

Credits FA
Lecture Practical Tutorial SA Total
FA1 FA2

02 02 - - 10 10 30 50
Prior knowledge of: Calculus, Linear Algebra, Probability Theory, and Python programming are essential.

Course Objectives:
This course aims at enabling students,
1. To learn the fundamentals of Neural Networks and their various types.
2. To explore Generative AI models like GANs, VAEs, and Transformers.
3. To analyze the limitations of traditional RNNs and LSTMs.
4. To discuss current trends and future directions in Generative AI research.

Course Outcomes:
After learning the course, the students should be able to:
1. Understand the evolution of AI and the significance of Deep Learning.
2. Apply recurrent neural networks-based algorithms.
3. Apply the principles of encoder-decoder architecture to implement GANs and VAEs.
4. Design practical solutions using advanced neural networks for diverse applications.

Detailed Syllabus

Unit Description Duration

[Hrs]
Foundations of AI and Neural Networks: History and evolution of AI/ML, Structure of Artificial
I Neural Networks, Steps in Training an Artificial Neural Network, Parameters and Hyperparameters, 08
Backpropagation.

Sequential Neural Network: History of natural language processing, feedforward, recurrent neural
II 08
networks, Seq2Seq architecture, and Long Short-Term Memory Networks (LSTMs).

Generative Models: Introduction to Encoder-Decoder Architecture and Introduction to Generative AI

III 07
Models, Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs).

Department of Multidisciplinary Studies MDM-9

 Page 5 of 17

Generative AI Applications: Applications in Various Fields: Art and Creativity, Image and Video
IV Generation, Text Generation, Music Composition, and Healthcare Finance. Real-world use cases and 07
challenges in deploying generative AI models.

Total 30

Text Books:
1. “Generative AI for everyone: Understanding the essentials and applications of this breakthrough technology"..
2. “Introduction to Generative AI”, Numa Dhamani, Kindle Edition, 2024.
3. “Neural Networks and Deep Learning: A Textbook” by Charu C. Aggarwal.

Reference Books:
1. "Generative Adversarial Networks Cookbook: Over 100 recipes to build generative models using Python, TensorFlow,
and Keras" by Josh Kalin.
2. “Generative AI in Software Development: Beyond the Limitations of Traditional Coding” Jesse Sprinter, 2024.

e-sources:
1. [Link]
s/?v=c86ee0d9d7ed
2. [Link]

Department of Multidisciplinary Studies MDM-9

 Page 6 of 17

MDM: B. Tech.: Civil/Comp/CompR/E&Tc/Mech/IT Semester: IV

Generative AI

Course: Generative AI Models Code: BCS24MD02

Teaching Scheme (Hrs./Week) Evaluation Scheme and Marks

Credits FA
Lecture Practical Tutorial SA Total
FA1 FA2

02 02 - - 10 10 30 50
Prior knowledge of: Calculus, Linear Algebra, Probability Theory, and Python programming, Introduction to
Generative AI, AI models are essential.
Course Objectives:
This course aims at enabling students,
1. Understand and apply the fundamentals of Generative AI and large language models.
2. Implement GPT architecture for text generation and dialogue systems.
3. Fine-tune BERT for downstream NLP tasks and explore advanced Transformer architectures.
4. Evaluate real-world applications, challenges, and emerging trends in Generative AI.

Course Outcomes:
After learning the course, the students will be able to:
1. Understand large language models' architecture and pre-training techniques.
2. Apply the GPT model for natural language processing tasks.
3. Apply the BERT model for natural language processing tasks.
4. Analyzing the performance and limitations of large language models.

Detailed Syllabus

Unit Duration
Description
[Hrs]
I Introduction to Large Language Models : 08
Overview of Generative AI and Large Language Models. Basics of attention mechanisms and
Transformer architecture. Pre-training techniques and transfer learning strategies.
II GPT Models and Applications: 08
Study of GPT architecture and variants. Applications of GPT models in text generation and
dialogue systems. Case study based implementation of GPT-based tasks. GPT-based chatbot
enhances E-Shop's customer support service

Department of Multidisciplinary Studies MDM-9

 Page 7 of 17

III BERT and Advanced Techniques: 07

Understanding BERT architecture and pre-training objectives. Fine-tuning BERT for downstream
NLP tasks. Exploration of advanced Transformer architectures and techniques
IV Applications and Future Directions: 07
Real-world applications of large language models. Challenges and limitations of current
approaches. Emerging trends and future directions in Generative AI.
Total 30

Text Books:
1. Generative AI for Everyone: Understanding the Essentials and Applications of This Breakthrough Technology". Altaf Rehmani
2. "Deep Learning" by Ian Goodfellow, Yoshua Bengio, and Aaron Courville.
3. "Neural Networks and Deep Learning: A Textbook" by Charu C. Aggarwal.
Reference Books:
1. "Generative Adversarial Networks Cookbook: Over 100 recipes to build generative models using Python, TensorFlow,
and Keras" by Josh Kalin.
2. “Generative AI in Software Development: Beyond the Limitations of Traditional Coding” Jesse Sprinter, 2024.

e-sources:
1. [Link]
s/?v=c86ee0d9d7ed
2. [Link]

Department of Multidisciplinary Studies MDM-9

 Page 8 of 17

MDM: B. Tech.: Civil/Comp/CompR/E&Tc/Mech/IT Semester: V

Generative AI

Course: Generative AI Tools & Techniques Code: BCS25MD03

Teaching Scheme (Hrs./Week) Evaluation Scheme and Marks

Credits FA
Lecture Practical Tutorial SA Total
FA1 FA2

03 03 - - 20 20 60 100
Prior knowledge of: Calculus, Linear Algebra, Probability Theory, and Python programming, Introduction to
Generative AI, AI models are essential.
Course Objectives:
This course aims at enabling students,
1. Develop proficiency in Python and TensorFlow 2 for preprocessing and cleaning data in
generative AI applications.
2. Implement and optimize GANs and VAEs for image generation, including training, fine-tuning, and
advanced techniques.
3. Master text generation techniques using LSTM and Transformer models, with a focus on fine-tuning
language models for various applications.
4. Explore music generation through different models like LSTM and Transformer, evaluating and fine-tuning
compositions for diverse music applications.

Course Outcomes:

After learning the course, the students should be able to:

1. Understand generative AI principles and word embeddings for text representation.
2. Apply Large Language Models effectively, including pre training and transfer learning.
3. Evaluate Generative AI models using appropriate metrics.
4. Implement Generative Adversarial Networks (GANs) for image generation tasks using
TensorFlow.
5. Identify different types of generative AI models suitable for music generation.
6. Analyze ethical consideration for music composition using GenAI

Detailed Syllabus

Department of Multidisciplinary Studies MDM-9

 Page 9 of 17

Unit Duration
Description
[Hrs]
I Python and TensorFlow 2 in Generative AI: 08
Overview of Python and TensorFlow 2, Preprocessing and cleaning data for Generative AI
applications. Visualizing data distributions and patterns in Generative AI datasets.
Introduction to TensorFlow's computation graph and eager execution
II Image Generation with Generative AI: 07
Introduction to Image Generation, Implementing GANs for Image Generation Training and Fine-
Tuning GANs , Generating Images with VAEs, Advanced Techniques in Image Generation, and
Image and Video Generation Applications.
III Text Generation with Generative AI: 08
Introduction to Text Generation, LSTM-based Text Generation, Transformer-based Text Generation,
Fine-Tuning Language Models, and Text Generation Applications
IV Music Generation with Generative AI: 07
Introduction to Music Generation, Music Representation, and LSTM-based Music Generation.
Transformer-based Music Generation, Evaluation and Fine-Tuning, Music CompositionApplications
MuseGAN:
V 08
Overview of MuseGAN architecture, Multi-track music generation using MuseGAN , Training
MuseGAN on polyphonic music datasets, Generating complex music compositions with MuseGAN
Advanced Techniques and Applications
VI 07
Transfer learning in music generation , Fine-tuning generative models for specific music genres or
styles , Ethical considerations in AI-generated music, Future directions and emerging trends in AI-
driven music composition.
Total 45

Text Books:
1. Generative AI with Python and TensorFlow 2: Create images, text, and music with VAEs, GANs, LSTMs, Transformer models” ,
Joseph Babcock and Raghav Bali , 2024
2. Generative AI for everyone: Understanding the essentials and applications of this breakthrough technology". Altaf Rehmani .
3. "Deep Learning" by Ian Goodfellow, Yoshua Bengio, and Aaron Courville.
4. "Neural Networks and Deep Learning: A Textbook" by Charu C. Aggarwal.
Reference Books:
1. "Generative Adversarial Networks Cookbook: Over 100 recipes to build generative models using Python, TensorFlow, and Keras"
by Josh Kalin.
2. “Generative AI in Software Development: Beyond the Limitations of Traditional Coding” Jesse Sprinter, 2024.

e-sources:
1. [Link]
courses/?v=c86ee0d9d7ed
2. [Link]

Department of Multidisciplinary Studies MDM-9

 Page 10 of 17

MDM: B. Tech.: Civil/Comp/CompR/E&Tc/Mech/IT Semester: V

Generative
AI

Course: Generative AI Tools and Techniques Laboratory Code: BCS25MD04

Teaching Scheme (Hrs. /Week) Evaluation Scheme and Marks

Credits Theory Practical Tutorial TW OR PR Total

01 - 02 - - 50 - 50
Prior knowledge: Python programming

Course Objectives:
1. To learn Python and TensorFlow skills for Generative AI.
2. To study techniques for cleaning and preparing data for Generative AI tasks.
3. To implement generative AI models
4. To develop innovative applications using generative AI tools and techniques.

Course Outcomes:
After learning the course, students will be able to:
1. Implement Python and TensorFlow basics, including data handling and preprocessing techniques.
2. Implement Generative AI models such as GANs, VAEs, LSTM networks, and Transformer models for
image, text, and music generation tasks.
3. Evaluate model performance and experiment with hyperparameters and optimization techniques to
enhance Generative AI outcomes.
4. Develop innovative applications in image, text, and music generation, showcasing practical skills.

Guidelines for Students:

1. The laboratory assignments are to be submitted by students in the form of a journal.
2. Each assignment write-up should have a title, objectives, outcomes, Theory- Concept in brief, dataset used,
data description, conclusion, and assessor's signature.
3. Program codes with sample output of all performed assignments should be submitted.

Guidelines for Laboratory /TW Assessment:

1. Continuous assessment of laboratory work is done based on the overall performance and laboratory
performance of the students.
2. Each laboratory assignment assessment should assign grade/marks based on parameters with appropriate
weightage.
3. Suggested parameters for overall assessment as well as each laboratory assignment assessment include- timely
completion, performance, innovation, efficiency, punctuality, and neatness.

Department of Multidisciplinary Studies MDM-9

 Page 11 of 17

Guidelines for Laboratory Conduction & Assessment

1. The suggested list of assignments is given below. Instructors can frame suitable assignments.
2. Perform any 7 assignments which include data preprocessing, GAN, text generation, image, video and music
Generation.
Detailed Syllabus

Assignment
No. Suggested List of Assignments

1 Write Python scripts to implement basic operations and TensorFlow 2 tensors.

2 Preprocess and clean datasets for Generative AI applications using Python libraries such as
Pandas and NumPy. Handle missing data, normalize features, and encode categorical variables.
3 Use Matplotlib or Seaborn to visualize data distributions and patterns in Generative AI datasets.
Plot histograms, scatter plots, and heatmaps to analyze data characteristics.
Implement a Generative Adversarial Network (GAN) architecture using TensorFlow 2. Train the
4
GAN model on a dataset such as MNIST or CIFAR-10 for image generation tasks.

5 Train a GAN model on a custom dataset for image generation. Experiment with
hyperparameters, loss functions, and optimization techniques to optimize GAN training.
Explore advanced techniques such as Wasserstein GANs, Progressive GANs, or StyleGANs for
6
image generation. Implement and compare these techniques for generating high-quality images.

7 Develop applications for image and video generation using trained Generative AI models. Use
the models to generate art, create deep fakes, or synthesize video content.
Text Generation: Implement a Long Short-Term Memory (LSTM) network using TensorFlow
8
2 for text generation tasks. Train the LSTM model on a dataset of text sequences and generate
new text samples.
Text generation: Implement a Transformer-based language model (e.g., GPT) using
9
TensorFlow 2 for text generation. Fine-tune the model on a text corpus and generate coherent
and contextually relevant text.
Text generation: Fine-tune a pre-trained language model (e.g., GPT, BERT) using transfer
10
learning techniques. Fine-tune the model on a domain-specific dataset and evaluate its
performance for text generation tasks.
11 Text generation: Develop applications for text generation tasks such as story generation,
dialogue generation, or code generation using trained Generative AI models.
Music Generation:
12
Preprocess music data and represent it in a suitable format for music generation tasks. Explore
MIDI or audio representations for training Generative AI models.
Music Generation:
13 Implement a Long Short-Term Memory (LSTM) network using TensorFlow 2 for music
[Link] the LSTM model on a dataset of music sequences and generate new musical
compositions.
Generate Novel Music Compositions:
14 Transformer-based Music Generation: Implement a Transformer-based architecture (e.g.,
MusicBERT, MusicGPT) using TensorFlow 2 for music generation. Fine-tune the model on a
music dataset and generate novel music compositions.

Department of Multidisciplinary Studies MDM-9

 Page 12 of 17

Text Books:
1. Responsible AI: Implementing Ethical and Unbiased Algorithms, by Shashin Mishra and Sray Agarwal
2. Generative AI in Practice: 100+ Amazing Ways Generative Artificial Intelligence is Changing Business and
Society, Bernard Marr

References:
1. “Generative AI with Python and TensorFlow 2: Create images, text, and music with VAEs, GANs, LSTMs,
Transformer models”, Joseph Babcock and Raghav Bali
2. "Generative Adversarial Networks: An Overview" by Vinod Nair and Geoffrey E. Hinton.
3. "Hands-On Generative Adversarial Networks with PyTorch 1.x" by Stefano Bosisio and Vijayabhaskar J.

Department of Multidisciplinary Studies MDM-9

 Page 13 of 17

MDM: B. Tech.: Civil/Comp/CompR/E&Tc/Mech/IT Semester: VI

Generative AI

Course: Generative AI Applications Code: BCS26MD05

Teaching Scheme (Hrs./Week) Evaluation Scheme and Marks

Credits FA
Lecture Practical Tutorial SA Total
FA1 FA2

02 02 - - 10 10 30 50
Prior knowledge of: Calculus, Linear Algebra, Probability Theory, and Python programming, are essential.

Course Objectives:
This course aims at enabling students,
1. To analyze the impact of GenAI in business
2. Understand ethical implications of Generative AI in science.
3. Analyze practical case studies for real-world insights.

Course Outcomes:
After learning the course, the students should be able to:
1. To understand the impact of generative AI in various fields.
2. Apply the concept of Gen AI in finance.
3. Apply Generative AI for simulation and modeling in various scientific fields.
4. Recognize the importance of ethical algorithms for fairness in AI systems.

Detailed Syllabus

Unit Duration
Description
[Hrs]

I Introduction 08
The Impact of Generative AI on Business and Society, Emerging Trends and Technologies in
Generative AI, Ethical Challenges and Regulatory Frameworks, Real-World Applications of
Large Language Models, and Challenges and Limitations of Current Approaches.
II Generative AI in Finance and Business 08
Algorithmic Trading and Financial Forecasting, Fraud Detection and Risk Management
Customer Service and Chatbots, Marketing and Advertising Creativity, Business Process
Automation and Optimization

Department of Multidisciplinary Studies MDM-9

 Page 14 of 17

Generative AI in Science and Research

III 07
Scientific Discovery and Hypothesis Generation
Simulation and Modeling in Various Fields, Data Augmentation and Synthesis, Collaboration
and Knowledge Sharing, and Ethical and Social Implications of Generative AI.
Responsible AI
IV 07
Importance of ethical and unbiased algorithms in AI systems, Ethical frameworks and principles
for AI development, Techniques for Implementing Unbiased Algorithms, Roles and responsibilities of
stakeholders in the AI ecosystem. Case Study : IBM's AI Fairness 360 Toolkit
Total 30

Text Books:
1. Generative AI in Practice: 100+ Amazing Ways Generative Artificial Intelligence is Changing Business and
Society, Bernard Marr
2. Responsible AI: Implementing Ethical and Unbiased Algorithms, by Shashin Mishra and Sray Agarwal
Reference Books:
1. "Generative Adversarial Networks Cookbook: Over 100 recipes to build generative models using Python,
TensorFlow, and Keras" by Josh Kalin.
2. Applications of Generative AI, Zhihan Lyu, Springer International Publishing, 2024.

e-sources:
1. [Link]
es/?v=c86ee0d9d7ed
2. [Link]

Department of Multidisciplinary Studies MDM-9

 Page 15 of 17

MDM: B. Tech.: Civil/Comp/CompR/E&Tc/Mech/IT Semester: VII

Generative
AI

Course: Capstone Project Code: BCS27MD06

Teaching Scheme (Hrs. /Week) Evaluation Scheme and Marks

Credits Theory Practical Tutorial TW OR PR Total

04 - 08 - 100 50 - 150
Prior knowledge: Generative AI Tools and Techniques is essential.

Course Objectives:
1. Understand the fundamental concepts and technologies involved in Generative AI development.
2. Study the principles of GenAI models applied to various interdisciplinary tasks.
3. Collaborate with cross-functional teams to design and deploy AI solutions that address specific
challenges and opportunities in the target industries

Course Outcomes:
After completion of this course, the students will be able to,
1. Plan, design, and implement a complex GenAI application.
2. Use GenAI models and tools to design the multidisciplinary project .
3. Utilize appropriate tools and technologies for multidisciplinary project development
4. Collaborate effectively within a team environment.

Guidelines:
Guidelines for Students:
The purpose of Capstone project work is to generate a concept and execute it methodically, utilizing the
knowledge acquired during college mostly to innovate or streamline processes. A cohort of undergraduate
students in their final year will engage in project work. The task entails assessing the viability of the project,
devising a project plan, examining current systems and available tools for project implementation, and
employing advanced software testing methods and technology using case tools. The objective is to transform
the design into a functional prototype.
Project work should be done in groups and each group formed in 3-4 students.
Self-sponsored project: The expenses incurred towards the completion of the project work will be borne by the
students.
Industry / Research institutes sponsored project: The sponsoring industry or research institute will provide
financial support for the expenses associated with completing the project work. Students are required to submit
a sponsorship letter or relevant paperwork that includes all the essential information such as the student's name,
guide's name, problem definition, work to be conducted, and sponsorship details.

Department of Multidisciplinary Studies MDM-9

 Page 16 of 17

Institute sponsored project: The expenses incurred towards the completion of the project work will be
supported by any of the institutes or organizations. Students shall submit the sponsorship letter or relevant
document mentioning the sponsorship in monetary support from the institutes or organizations.
Guidelines for Assessment:
1. The project work assessment will be done in following stages:
Review 1: Starting of the semester (within 15 Days) [Presentation & synopsis]
Review 2: After 1 month of review 1 [Presentation]
Review 3: After 1 month of review 2 [Presentation and Report]
Final Examination: At the end of semester

Detailed Syllabus

Suggested List of projects

Manufacturing :

1 Gen AI based quality control system to identify defects in manufactured products and improve
overall product quality.
Develop a generative AI model to identify and classify defects in manufactured products, such as
2
surface imperfections, dimensional inaccuracies, or material inconsistencies, to improve quality
control processes in manufacturing.
Create a generative AI tool that generates customized product designs based on customer
3
preferences and specifications, enabling mass customization and personalized manufacturing
solutions.
Develop a generative AI predictive maintenance system that analyzes equipment sensor data to
4
forecast potential failures, schedule proactive maintenance tasks, and optimize equipment
performance in manufacturing facilities.
Construction :
5 Optimize construction schedules, automate tasks, and improve safety.

6 Generate Construction blueprints as per clients specification.

Develop a generative AI system that can automatically generate building designs based on
7
project requirements, site constraints, and architectural preferences, streamlining the design
process and enabling rapid prototyping of construction projects.
Create a generative AI tool that can recommend optimal material selections for construction
8
projects based on factors such as cost, durability, sustainability, and performance, helping
architects and engineers make informed decisions during the design phase.
Electronics & Telecommunication :
Developing a predictive maintenance system using generative AI for telecommunications
9
equipment.

Develop a generative AI system to optimize signal processing algorithms for

10
telecommunications applications, such as noise reduction, channel equalization, and data
compression, to improve signal quality and transmission efficiency.

Department of Multidisciplinary Studies MDM-9

 Page 17 of 17

Implement a generative AI model to optimize antenna designs for wireless communication

11
systems, satellite communication, or radar applications, aiming to enhance signal coverage,
bandwidth efficiency, and overall performance.
Create a generative AI tool that can predict network traffic patterns and bandwidth demands in
12
telecommunications networks, enabling proactive network management, resource allocation, and
capacity planning to ensure optimal performance.

Department of Multidisciplinary Studies MDM-9