This advanced generative AI course focuses on diffusion models, one of the most powerful modern techniques used in image generation and deep learning systems. It is designed for learners who already understand basic machine learning and want to dive into state-of-the-art generative modeling.

The course begins with an introduction to diffusion models and their mathematical foundations. It explains how data is gradually transformed through noise addition and then reconstructed through a reverse process, which is the core idea behind modern image generation systems.

You will study key concepts such as score-based models, Langevin dynamics, and DDPM (Denoising Diffusion Probabilistic Models). These methods are essential for understanding how models like Stable Diffusion generate high-quality images.

The course also explores acceleration techniques that improve model efficiency and performance, making diffusion models faster and more practical for real-world applications.

Advanced topics include guided diffusion, inverse problems, and SDE/ODE interpretations, which connect diffusion models to continuous mathematical systems used in physics and probability theory.

In addition, student-led lectures provide deeper insights into specialized topics and research perspectives.

By the end of this course, you will have a strong theoretical and practical understanding of diffusion models and how they are used in modern generative AI systems for image and data generation.