Lightnews — Scholar-powered news

Rémi Flamary @rflamary.bsky.social · 4d

So because you don't want large batches you solve the whole OT problem! Love it.

Reposted by Rémi Flamary

David Picard @davidpicard.bsky.social · 5d

It's a great discussion because we all have different interpretation of what this means and what the consequences are.

Myself, I have absolutely no doubt that scaling works. If you have all the videos in the world and are able to train a model that can recall and merge any of them, then for sure...

1 4 12

Rémi Flamary @rflamary.bsky.social · 5d

This is why I love bsky! Totally agree with you. An important question of science is why and how is it working. This is what we should look at and promote as a community.

6

Rémi Flamary @rflamary.bsky.social · 6d

FOMO is our careers nowadays.

3

Rémi Flamary @rflamary.bsky.social · 7d

OK we should probably talk directly ;).

1 1

Rémi Flamary @rflamary.bsky.social · 7d

And yes I already checked the .h file it is very simple, so for POT we need a cython wrapper and a way to allows wheels compilation with eigen without exploding (when compiled from sources) but this would be very nice.

Rémi Flamary @rflamary.bsky.social · 7d

That was exactly my question to @nbonneel.bsky.social 10 se ago! I would erally like to have larger scale solvers even for low dims. I know that in some physics applications (OT on particles) scaling was a problem

1

Rémi Flamary @rflamary.bsky.social · 7d

Also I was wondering how well it works in higher dimensions (up to few hundreds) when you said you don't use the Gaussian slicing? Errors wrt true solver as a function of d and k would be very interesting

2

Rémi Flamary @rflamary.bsky.social · 7d

Hello, I'd be interested to have a Python/numpy wrapper of this awesome solver in POT. Would you be interested in helping out?

2 5

Rémi Flamary @rflamary.bsky.social · 8d

Figure 1. Happy ML researcher and open source developer presenting his toolbox SKADA at PyData Paris. Congrats @tgnassou.bsky.social the presentation was awesome!

1 9

Reposted by Rémi Flamary

Sonia Mazelet @soniamazelet.bsky.social · 9d

I am happy to share that our paper "Unsupervised Learning for Optimal Transport plan prediction between unbalanced graphs" was accepted at Neurips 2025 ! 🥳

Huge thanks to my co-authors @rflamary.bsky.social and Bertrand Thirion !

arxiv.org/abs/2506.12025

(1/5)

Unsupervised Learning for Optimal Transport plan prediction between unbalanced graphs

Optimal transport between graphs, based on Gromov-Wasserstein and other extensions, is a powerful tool for comparing and aligning graph structures. However, solving the associated non-convex optimizat...

arxiv.org

1 4 29

Rémi Flamary @rflamary.bsky.social · 10d

The demon princes is very early Jack Vance. I really love his more recent books such as the Alastor trilogy that is basically in the same universe and is much more original.

1

Rémi Flamary @rflamary.bsky.social · 14d

We should apply Queueing Theory there. Pretty sure that there is a diverging amount of good paper in the queue.

2

Rémi Flamary @rflamary.bsky.social · 14d

At least 500 more that were rejected by neurips for no scientific reasons I guess.

1

Rémi Flamary @rflamary.bsky.social · 14d

Did not know that all authors that submit 3 ICLR of more must be reviewers with 6 (or more?) papers. Seems fair to review 2 papers per submission but after the full NeurIPS summer (and rejection) I would have appreciated a reduced load to do them better (I say that as an ICML 2025 top reviewer).

1 2 11

Rémi Flamary @rflamary.bsky.social · 15d

I missed it last month! Very happy to present POT next week at PyData Paris.

PyData Paris @pydataparis.bsky.social · Jul 29

📢 Talk Announcement

"Optimal Transport in Python: A Practical Introduction with POT", by Rémi Flamary.

📜 Talk info: pretalx.com/pydata-paris-2025/talk/FNCKXU
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets

2 9

Reposted by Rémi Flamary

PyData Paris @pydataparis.bsky.social · Jul 29

📢 Talk Announcement

"Optimal Transport in Python: A Practical Introduction with POT", by Rémi Flamary.

📜 Talk info: pretalx.com/pydata-paris-2025/talk/FNCKXU
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets

2 6

Rémi Flamary @rflamary.bsky.social · 19d

SKADA is a beautiful software for Domain Adaptation in python with many shallow and deep methods implemented, it is 100% compatible with @scikit-learn.org models and pipelines and with @pytorch.org for deep learning methods.

PyData Paris @pydataparis.bsky.social · 19d

📢 Talk Announcement

"Tackling Domain Shift with SKADA: A Hands-On Guide to Domain Adaptation", by Théo Gnassounou and Antoine Collas.

📜 Talk info: pretalx.com/pydata-paris-2025/talk/MEPTCE
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets

1 9

Reposted by Rémi Flamary

PyData Paris @pydataparis.bsky.social · 19d

📢 Talk Announcement

"Tackling Domain Shift with SKADA: A Hands-On Guide to Domain Adaptation", by Théo Gnassounou and Antoine Collas.

📜 Talk info: pretalx.com/pydata-paris-2025/talk/MEPTCE
📅 Schedule: pydata.org/paris2025/schedule
🎟 Tickets: pydata.org/paris2025/tickets

1 2

Reposted by Rémi Flamary

Bruno Levy @brunolevy01.bsky.social · 27d

Our article "Large-scale semi-discrete optimal transport with distributed Voronoi diagrams" (with Nicolas Ray, Quentin Merigot and Hugo Leclerc) just got accepted by Journal of Computational Physics. Pushing the limits of Optimal Transport by several orders of magnitude !

A cosmological simulation, showing the cosmic web, with 300 million galaxy clusters. A nice test case for the new algorithm !

1 4 33

Reposted by Rémi Flamary

Anna Korba @akorba.bsky.social · Sep 8

I'm thrilled to announce that my #ERCStG project **Optinfinite : Efficient infinite-dimensional optimization over measures**
has been accepted. Thank you
@erc.europa.eu !
Many thanks also to @crestumr.bsky.social @ipparis.bsky.social for they support, as well as to my collaborators and friends.

4 5 16

Reposted by Rémi Flamary

Clément Canonne @ccanonne.github.io · Sep 4

To: Reviewer 2

My name is Inigo Montoya
You killed my paper
Prepare to die

1 6 54

Rémi Flamary @rflamary.bsky.social · Sep 3

Bea Wolf was the perfect oddball book. We had tremendous fun reading it with my kid and it left its mark. She still finds people in real world and comics that look like some obscure character from the book that like a stupid adult I forgot about.

1 1 1

Reposted by Rémi Flamary

Martin Trapp @trappmartin.bsky.social · Aug 29

Thanks for raising this issue. As someone on the authors' and reviewers' side, this feels like a violation of the reviewing process. One might wonder, “Why do we put so much time and effort into the peer-review process if decisions are ultimately made arbitrarily?”.

3 18

Reposted by Rémi Flamary

Ben Recht @beenwrekt.bsky.social · Aug 28

I wrote about the tragic death of Adam Raine and the venal negligence of "AI Safety." www.argmin.net/p/the-banal-...

The Banal Evil of AI Safety

Chatbot companies are harmful and dishonest. How can we hold them accountable?

www.argmin.net

1 39 76