How I Understand Flow Matching

Подписаться 4,3 тыс.

Просмотров 10 тыс.

50% 1

Flow matching is a new generative modeling method that combines the advantages of Continuous Normalising Flows (CNFs) and Diffusion Models (DMs).
In this tutorial, I share my understanding of the basics of flow matching and provide an overview of how these ideas evolve over time.
Check out the resources below to learn more about this topic.
===== Paper/blog survey =====
[Papamakarios et al. 2021] Normalizing flows for probabilistic modeling and inference arxiv.org/abs/...
[Kobyzev et al. 2020] Normalizing Flows: An Introduction and Review of Current Methods arxiv.org/abs/...
[Tor Fjelde et al. 2024] An Introduction to Flow Matching
mlg.eng.cam.ac...
[Jakub Tomczak] Flow Matching: Matching flows instead of scores
Blog: jmtomczak.gith...
Code example: github.com/jmt...
===== Research talks =====
[Yaron Lipman] Flow Matching: Simplifying and Generalizing Diffusion Models
• Flow Matching: Simplif...
[Michael S Albergo] Building Normalizing Flows with Stochastic Interpolants
• Building Normalizing F...
[Alex Tong] Conditional Flow Matching
• TransferLab Seminar: C...
Thumbnail background image credit: unsplash.com/p...

Опубликовано:

2 окт 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 44

@sayakpaul3152 2 месяца назад

Do you have the next video already? It's so good!

@jbhuang0604 2 месяца назад

Thanks! I am working on it. :-)

@Neo-kx3fe 3 месяца назад

@10:55 with that local outgoingness on the left, why there is one additional term p_t(x_t) inside the d/dx bracket? This term seems to disappear in @11:18. Thanks.

@manuellecha 4 месяца назад

Thank you very much! You did a really nice job! The video is clear, visual, and informative. It is consistent with the timeline and evolution of the field, and it effectively conveys the information along with the motivation for the development of these models.

@jbhuang0604 4 месяца назад

Glad you liked it! It’s a lot of fun making this video!

@kvu207 4 месяца назад

beautiful! May I ask how you made the animations for this video?

@jbhuang0604 4 месяца назад

Most of the animations are from the “morph transition” in PowerPoint slides. The rest are from Adobe premier pro.

@pravin1390 Месяц назад

Thanks for the explanation. I feel it makes sense to use the notation p(x_t, t). Then, it is clear to prove ∂/∂t p(x_t, t) = -div(p(x_t, t) u(x_t, t)) is equivalent to d/dt p(x_t, t) = -p(x_t, t) div(u(x_t,t)), which is indeed required here. Please let me know if my comments make sense. The first equation is partial derivative and second equation is total derivative is what I sense.

@julienblanchon6082 4 месяца назад

This is brilliant !

@jbhuang0604 4 месяца назад

Glad that you enjoyed the video!

@plcrodrigues 3 месяца назад

This is gold. Congrats for doing such a fun, pedagogical, and informative on a topic that can often be quite dry in the literature.

@jbhuang0604 3 месяца назад

Thanks for your kind words!

@AnujZore-pe9mg 3 месяца назад

Thanks once again for easy-to-understand explanation! Gonna miss CSMC 733 lectures :(

@jbhuang0604 3 месяца назад

Glad you like them!

@因幡の黒うさぎ-i1p 5 дней назад

Your lecture was truly inspiring and made complex concepts so easy to understand. Thank you for your incredible clarity and passion - I’m deeply grateful!

@jbhuang0604 День назад

You're very welcome! Glad that you like it!

@ruoshiliu6024 3 месяца назад

Amazing work Jia-Bin!! P.S. you should create a bibtex for this video so it can be cited in literature :P

@jbhuang0604 3 месяца назад

Haha! Thanks! Too bad google scholar don’t include views of RU-vid videos.

@JQXU-z3s 2 месяца назад

Thank you for your excellent work! Absolutely clear and informative.

@jbhuang0604 2 месяца назад

Glad it was helpful!

@morrisfan2004 28 дней назад

I wondering if there are any urgent, potential applications of flow matching in industry?

@jbhuang0604 День назад

I think many of the recent text-to-image generation models are now trained with flow matching. There are also many other applications beyond image generations.

@r00t257 4 месяца назад

Legend comeback 🙇! Your educational video is worth more than gold.💓🙏

@jbhuang0604 4 месяца назад

Thanks a lot! Glad you like it!

@nathan_ca 3 месяца назад

Thanks! This is amazing video to get students, like me, to re-engage with these topics that i haven't had a chance to explore more ❤

@jbhuang0604 3 месяца назад

Thanks! Glad that this is helpful.

@rexton123 2 месяца назад

What are the advantage of flow matching compared to diffusion models?

@jbhuang0604 День назад

You can view it as a generalization of diffusion models. The training can converge faster and you won't have the difficulty where you cannot reach pure Gaussian distributions using finite diffusion steps.