Course Overview
Generative modeling is a critical area in machine learning that focuses on learning data distributions and generating new, realistic data samples. In recent years, diffusion models and flow matching models have emerged as powerful approaches for building high-quality generative models, demonstrating success in image generation, text-to-image synthesis, and various other domains.
This course provides a deep dive into the theory, methodology, and practical applications of diffusion models and flow matching models, as well as how they compare to other generative modeling techniques. Students will learn to implement and experiment with these models, understand their underlying principles, and apply them to real-world tasks such as image generation, superresolution, and generating protein ensembles. The students will also explore intriguing open problems within the field of generative modeling and be encouraged to develop innovative solutions and new perspectives.
- Lecture Times: Tuesdays & Thursdays, 3:30 PM - 4:50 PM
- Instructor: Zhizhen Zhao
- Location: Online, Zoom link
- Office Hour: 45 minutes after each lecture
- TA: Yichi Zhang
- Recorded Lectures are available at media space
- Campuswire enrollment: link
Schedule
| Event | Date | Topic | Notes |
|---|---|---|---|
| Lecture 1 | 08/26/2025 | Introduction & syllabus | Slides |
| Lecture 2 | 08/28/2025 | Mathematical foundations for generative modeling | Slides |
| Lecture 3 | 09/02/2025 | Overview of generative models I | Slides |
| Lecture 4 | 09/04/2025 | Overview of generative models II | Slides |
| Lecture 5 | 09/09/2025 | Continuous normalizing flow and Neural ODE | |
| Lecture 6 | 09/11/2025 | Energy based model and score matching | |
| Lecture 7 | 09/16/2025 | Intro to diffusion models | |
| Lecture 8 | 09/18/2025 | Diffusion models II | |
| Lecture 9 | 09/23/2025 | Design space of diffusion models | |
| Lecture 10 | 09/25/2025 | Consistency models | |
| Lecture 11 | 09/30/2025 | Masked diffusion models | |
| Lecture 12 | 10/02/2025 | Rectified flow and flow matching | |
| Lecture 13 | 10/07/2025 | Stochastic interpolants and Schrodinger bridge | |
| Lecture 14 | 10/09/2025 | Flow matching with mini-batch coupling | |
| Lecture 15 | 10/14/2025 | Flow matching on general geometry | |
| Lecture 16 | 10/16/2025 | Flow matching with discrete state space | |
| Lecture 17 | 10/21/2025 | Connections with optimal transport | |
| Lecture 18 | 10/23/2025 | Generalization properties of diffusion and flow models—Empirical observation | |
| Lecture 19 | 10/28/2025 | Generalization properties of diffusion and flow models—Theoretical analysis | |
| Lecture 20 | 10/30/2025 | Low-dimensional adaptation and representation learning | |
| Lecture 21 | 11/04/2025 | Application: Imaging inverse problems | |
| Lecture 22 | 11/06/2025 | Diffusion posterior sampling and related methods | |
| Lecture 23 | 11/11/2025 | Application in video generation | |
| Lecture 24 | 11/13/2025 | Probablistic forecasting | |
| Lecture 25 | 11/18/2025 | Diffusion language models | |
| Lecture 26 | 11/20/2025 | Generative AI in molecular and protein sciences | |
| Break | 11/25/2025 | No Class - Thanksgiving | |
| Break | 11/27/2025 | No Class - Thanksgiving | |
| Lecture 27 | 12/02/2025 | Review and discussion | |
| Lecture 28 | 12/04/2025 | Project presentation | |
| Lecture 29 | 12/09/2025 | Project presentation | |
Assignments
- Homework 1
- Homework 2
- Literature review
- Project