Scribd
Upload
36 views3 pages

CSE-848, Generative AI and Applications

Uploaded by

mwattoo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
36 views3 pages

CSE-848, Generative AI and Applications

Uploaded by

mwattoo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

School of Interdisciplinary Engineering and Sciences (SINES)

National University of Sciences & Technology (NUST)

Course Title: Generative AI and Applications


Course Code: CSE-848
Pre-requisite: Atleast one course on Machine Learning /
Deep Learning
Course description:
The goal of the course on generative artificial intelligence is to impart advanced
knowledge and abilities in the subject. Machine learning algorithms are used in
generative AI to create and synthesize new data, including text and images. Students
who take this course will be given the skills they need to comprehend cutting-edge
generative models and develop their own by covering both the theoretical
underpinnings and practical applications of generative AI.

Recommended Books:
1. Foster D. Generative Deep Learning: Teaching Machines to Paint. Write,
Compose, and Play O’Reilly Media Incorporated. 2019:139-40.
2. Goodfellow I, Bengio Y, Courville A. Deep learning. MIT press; 2016 Nov 10.
3. Foster D. Generative deep learning. " O'Reilly Media, Inc."; 2022 Jun 28.
4. Tomczak JM. Deep generative modeling. Springer; 2022.
5. Babcock J, Bali R. Generative AI with Python and TensorFlow 2: Create
images, text, and music with VAEs, GANs, LSTMs, Transformer models. Packt
Publishing Ltd; 2021 Apr 30.
6. Manaswi NK. Generative Adversarial Networks with Industrial Use Cases:
Learning How to Build GAN Applications for Retail, Healthcare, Telecom, Media,
Education, and HRTech. BPB Publications; 2020 Mar 4.
7. van der Zant T, Kouw M, Schomaker L. Generative artificial intelligence.
Springer Berlin Heidelberg; 2013.
8. Ahirwar K. Generative adversarial networks projects: Build next-generation
generative models using TensorFlow and Keras. Packt Publishing Ltd; 2019 Jan
31.

Prerequisite: None
Objectives:
The educational objectives of the course “Generative AI and Applications” are as below:
a. Be familiar with the foundational concepts of generative AI, such as deep
learning, probabilistic modelling, and generative models. Learn about
various generative modelling approaches, including flow-based models,
generative adversarial networks (GANs), variational autoencoders
(VAEs), and autoencoders.
b. Learn advanced techniques for training and evaluating generative
models, including optimization methods, loss functions, and evaluation
metrics.
c. Gain insights into the ethical considerations and challenges associated
with generative AI, including issues related to bias, fairness, and privacy.
d. Develop practical skills in implementing and fine-tuning generative
models using popular deep learning frameworks, such as TensorFlow
and PyTorch.
e. Apply generative AI techniques to various domains, such as image
synthesis, text generation, music composition, and data augmentation.
f. Explore state-of-the-art research papers and advances in generative AI,
and critically analyze their implications and potential applications.
g. Undertake a hands-on project that involves designing, implementing,
and evaluating a generative AI system in a chosen domain.
Course Learning Outcomes:
At the end of the course, the students should be able to:
1. Understand the fundamental ideas of generative AI, such as deep learning,
probabilistic modelling, and generative models.
2. Learn about advanced training and assessment methods for generative
models, such as optimisation strategies, loss functions, and evaluation
metrics.
3. Learn about the ethical dilemmas and difficulties raised by generative AI, such
as those with bias, justice, and privacy.
4. Learn how to create and optimise generative models using well-known deep
learning frameworks.
5. Apply generative AI methods to a variety of fields, including data
augmentation, text generation and image synthesis.
Assessment system:
Quizzes 10-15%

Assignments 5-10%

MSE 30-40%

ESE 40-50%
Updated Curriculum- MS/PhD Computational Science & Engineering Program Annex- H

Week wise Lecture Plan:


Week Lecture Topic Assignments Quizzes
1 Introduction to Generative AI
2  Introduction to generative models
 Scenarios that are challenging and require
Generative AI
 Probability theory and generative modeling
 Deep learning foundations for generative AI
3  Normalizing Flow Models
4 Autoencoders and Variational Autoencoders Assignment 1
(VAEs)
5  Autoregressive Models
 Variational autoencoders
7 Generative Adversarial Networks (GANs) Quiz 1
 Introduction to GANs and their components
8  Training GANs and improving stability
 Conditional and progressive GANs
9 Transformers and Language Models Quiz 2
 Transformers
10  Text generation and language modeling
11  Large Language Models Assignment 2
12 MSE
13  Optimization methods for generative models
 Evaluation Metrics for Generative Models
14  Data augmentation using generative models
 Prompt programming and neural text Assignment 3
decoding
16 Ethical Considerations in Generative AI
 Bias and fairness in generative models Quiz 3
 Privacy implications and considerations
 Addressing ethical challenges in generative
AI
17 Advances in Generative AI
 Review of recent research papers and
breakthroughs
 Exploring cutting-edge architectures and
techniques
 Emerging trends and future directions in
generative AI
18 ESE

81
WP No. 62-66th ACM-1 Sep 2023

You might also like