AI/ML STIG Lecture Series
Artificial Intelligence and Machine Learning Science and Technology Interest Group (AI/ML STIG)
Location
Virtual
Dates
23 March 2026
4:00pm ET
Community
AI/ML STIG
Type
Seminar
Diffusion Models
Speaker
Duo Xu, U. of Toronto
An introduction to denoising diffusion probabilistic models (DDPMs) for astronomy, covering the mathematical foundations of forward and reverse diffusion processes, conditional generation of synthetic galaxy images, and image-to-image translation for data restoration.
Topics Covered
- Motivation: from data-starved to data-dominated astronomy
- Generative model landscape: VAEs, GANs, normalizing flows, diffusion, flow matching, and score-based models
- Forward and reverse diffusion processes
- Conditional DDPMs for synthetic galaxy generation
- Image-to-image translation for astronomical data restoration
Session Recording
Meeting Connection
News Straight to Your Inbox
Subscribe to your community email news list
We will never share your email address.



