eupskills | Foundations and Applications of Generative AI

Course / Course Details

Foundations and Applications of Generative AI

Generative AI

By - Sateesh
0 students
6 Hours
(0)

Course Requirements

Proficiency in programming, particularly in Python

Knowledge of machine learning fundamentals, including basic supervised and unsupervised learning

Understanding of deep learning principles (e.g., neural networks, backpropagation)

Familiarity with linear algebra, calculus, probability, and statistics

Experience with data processing and analysis tools such as TensorFlow or PyTorch is beneficial but not mandatory

Course Description

This course dives into the rapidly evolving field of Generative AI, exploring how machines create content such as text, images, audio, and more. Starting from the basics, students will study the architecture and algorithms behind models like Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and transformer-based models (e.g., GPT, BERT). Emphasis will be on understanding how generative models learn patterns from data to produce realistic outputs and developing hands-on skills to build and fine-tune these models. Topics will also include the ethical and societal impacts of generative AI, as well as best practices for deploying these models responsibly.

Course Outcomes

By the end of this course, students will be able to:

Understand core concepts and mechanisms of generative models, including GANs, VAEs, and transformer-based architectures.
Build, train, and evaluate generative AI models for various applications, such as text generation, image synthesis, and style transfer.
Utilize advanced neural network techniques to improve model performance and generate high-quality outputs.
Develop custom generative AI solutions by applying transfer learning, fine-tuning, and optimizing model parameters.
Identify and address ethical considerations, biases, and potential misuse of generative models in real-world applications.
Gain practical experience with state-of-the-art frameworks (e.g., TensorFlow, PyTorch) to implement and deploy generative AI models in production environments.

Course Curriculum

2 chapters
3 lectures
0 quizzes
6 Hours total length

Toggle all chapters

1 Introduction to Generative Models

Preview

1 Hour

Overview of generative AI and its applications Types of generative models: probabilistic vs. deep learning-based Basics of data generation and synthetic data

2 Variational Autoencoders (VAEs)

Preview

40 Min

Introduction to VAEs: Explanation of VAEs and how they differ from GANs in approach and application, with a focus on probabilistic modeling. Encoder-Decoder Architecture: Deep dive into the encoder-decoder structure of VAEs and the concept of latent space representation for generating data. Building a VAE Model: Practical session on constructing a basic VAE, with code and examples to understand the model’s workflow from encoding input data to reconstructing and generating new data

1 Generative Adversarial Networks (GANs)

Preview

30 Min

Overview of GANs: Introduction to GANs and the concept of adversarial training, where two neural networks, a generator and a discriminator, compete in a zero-sum game. Structure of GANs: Detailed breakdown of the discriminator and generator networks, explaining their roles in generating and validating realistic data samples. Implementing a Basic GAN: Step-by-step guide to building a simple GAN for image generation, covering essential components and code examples.

1. Presentation

Download

Instructor

Sateesh

Sateesh A, bringing extensive expertise in AI-driven healthcare solutions. With a background in Engineering (B.E., M.Tech., and Ph.D. in AI for Healthcare), Sateesh has over two decades of experience in technology innovation, spanning roles in top organizations like Samsung, Intel, and Sandisk.

His work focuses on developing advanced AI models for healthcare, including federated learning frameworks, synthetic data generation, and explainable AI models for predictive healthcare applications.

Sateesh has led multiple groundbreaking research projects, authored numerous publications, and presented at international conferences, advancing the field of adaptive, teacher-less education platforms. His vision is to make high-quality education accessible to remote regions through AI, helping learners achieve employability and lifelong learning.

Sateesh’s achievements include recognition as a top tech innovator, participation in elite accelerator programs, and collaboration with esteemed institutions like Stanford, IIIT-B, and IIM-A.

0 Rating