Course Description

This tutorial covers optimization strategies for diffusion models used in image and video generation; topics include architecture design, hardware efficiency, and inference techniques. Speaker: Sayak Paul, Hugging Face.

This tutorial is part of a seminar series led by graduate students and postdocs in the MIT …

This tutorial covers optimization strategies for diffusion models used in image and video generation; topics include architecture design, hardware efficiency, and inference techniques. Speaker: Sayak Paul, Hugging Face.

This tutorial is part of a seminar series led by graduate students and postdocs in the MIT Department of Brain and Cognitive Sciences (BCS) from 2015 to the present, featuring topics relevant to research on intelligence in neuroscience, cognitive science, and artificial intelligence. The tutorials are aimed at participants who have some computational background but are not experts on these topics.

This series was organized by Sol Markman and Ajani Stewart, graduate students in the Department of Brain and Cognitive Sciences.

Course Info

Learning Resource Types

Tutorial Videos
Diagram showing a cat image turning into noise in four steps in the forward process, then noise turning back into the cat image in the reverse process.
Forward and reverse diffusion: design and optimization improvements result in a dramatic leap in text-to-image quality from early attempts to today’s results. (Figure by Dongqi Zheng. Source: arXiv. License: CC BY 4.0.)
MIT Open Learning
Over 2,500 courses & materials
Freely sharing knowledge with learners and educators around the world. Learn more
© 2001–2026 Massachusetts Institute of Technology
Accessibility
Creative Commons License
Terms and Conditions
Proud member of:
© 2001–2026 Massachusetts Institute of Technology