Getting Started with Diffusion Models in JAX | PyData Boston 🌟
Learn the fundamentals of diffusion models with Axel Donath in this beginner-friendly tutorial using JAX. Perfect for newcomers eager to explore cutting-edge generative AI techniques. Access the tutorial materials here: https://axeldonath.com/jax-diffusio
PyData
960 views • Jul 8, 2025
About this video
Axel Donath (https://axeldonath.com/)
See the tutorial materials here: https://axeldonath.com/jax-diffusion-models-pydata-boston-2025/
www.pydata.org
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases.
[00:00:00] - Welcome and speaker intro
[00:03:32] - What are diffusion models?
[00:05:56] - What is JAX? Quick tour
[00:10:03] - Physical diffusion and a first JAX example
[00:22:00] - Reverse diffusion with Langevin sampling and the score function
[00:36:05] - Score matching and an MLP with Equinox
[00:42:40] - From score matching to DDPM
[00:57:59] - Results: time-dependent scores and sampling
[01:05:10] - Q&A: JAX vs PyTorch and more
Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/numfocus/YouTubeVideoTimestamps
See the tutorial materials here: https://axeldonath.com/jax-diffusion-models-pydata-boston-2025/
www.pydata.org
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases.
[00:00:00] - Welcome and speaker intro
[00:03:32] - What are diffusion models?
[00:05:56] - What is JAX? Quick tour
[00:10:03] - Physical diffusion and a first JAX example
[00:22:00] - Reverse diffusion with Langevin sampling and the score function
[00:36:05] - Score matching and an MLP with Equinox
[00:42:40] - From score matching to DDPM
[00:57:59] - Results: time-dependent scores and sampling
[01:05:10] - Q&A: JAX vs PyTorch and more
Want to help add timestamps to our YouTube videos to help with discoverability? Find out more here: https://github.com/numfocus/YouTubeVideoTimestamps
Tags and Topics
Browse our collection to discover more content in these categories.
Video Information
Views
960
Likes
27
Duration
01:15:52
Published
Jul 8, 2025
Related Trending Topics
LIVE TRENDSRelated trending topics. Click any trend to explore more videos.
Trending Now