Samuel Vaiter retweeted
The alpha version of my new book "Optimal Transport
for Machine Learners" is out, with in particular an online version with interactive figures
gpeyre.com/ot4ml/
Samuel Vaiter retweeted
🏅Finalist: Gradient Scarcity in Graph Learning with Bilevel Optimization, by Hashem Ghanem, @vaiter, @n_keriven 5/n
openreview.net/forum?id=10YJ…
Samuel Vaiter retweeted
Happy to be at #NeurIPS2025 in San Diego to present our poster ‘Learning Theory for Kernel Bilevel Optimization’ #3005, Fri at 4:30 p.m. Stop by/ping me to chat, especially about statistics, causality, generative models! Let's connect!
Joint w/ E. Pauwels, @vaiter, @MichaelArbel
Samuel Vaiter retweeted
I'm BEYOND THRILLED to tell you that our first original song is now available on all platforms!
Plz share if you like it 🙏🏼🙏🏼😊
Cheers
Samuel Vaiter retweeted
🚀 NeurIPS@Paris is back for a 5th edition at the SCAI. Meet us in central Paris to discuss recent advances in AI!
📆 25th & 26th Nov. 2025
🌐neuripsinparis.github.io/neu…
🎓 Committee: Chloé-Agathe Azencott, @BachFrancis, Claire Boyer, @gerardbiau , @VianneyPerchet , @jeanphi_vert
Samuel Vaiter retweeted
If you are in the Bay Area, consider attending our workshop, "Graph Learning Meets TCS," at the Simons Institute (simons.berkeley.edu/workshop…).
Samuel Vaiter retweeted
New paper on the generalization of Flow Matching arxiv.org/abs/2506.03719
🤯 Why does flow matching generalize? Did you know that the flow matching target you're trying to learn **can only generate training points**?
with @Qu3ntinB, Anne Gagneux & Rémi Emonet 👇👇👇
Samuel Vaiter retweeted
I wrote a summary of the main ingredients of the neat proof by Hugo Lavenant that diffusion models do not generally define optimal transport. github.com/mathematical-tour…
Samuel Vaiter retweeted
Our @TmlrOrg paper "Gradient scarcity with Bilevel Optimization for Graph Learning" (w/ H Ghanem,
@vaiter ) was accepted as an oral presentation at @LogConference 🤓 100% free and online, come check it out!
arxiv.org/abs/2303.13964
Samuel Vaiter retweeted
I have several internship/PhD positions on Graph ML available with ERC MALAGA 🤓 See details here:
nkeriven.github.io/malaga/
Don't hesitate to contact me! (same if you are interested in a post-doc related to graph ML)
I'm thrilled to announce that my #ERCStG project has been accepted 🤓
**MALAGA: Reinventing the Theory of Machine Learning on Large Graphs**
Many job openings coming up, see nkeriven.github.io/malaga for updates! Thank you @ERC_Research and all my collaborators past and future
Samuel Vaiter retweeted
[Master Internship] With Paul Escande (CNRS, Toulouse), we are looking for an intern to work on the concentration of emprirical risk minimizers. iutzeler.org/assets/docs/Int…
Samuel Vaiter retweeted
How fast is gradient descent, *for real*?
Some (partial) answers in this new blog post on scaling laws for optimization.
francisbach.com/scaling-laws…
Samuel Vaiter retweeted
The story of my (academic) life ! Started on multi-image super-resolution, now working on the theoretical side of inverse problems.
BTW we have a paper with @RemiGribonval and @vaiter on the choice of optimal "atoms" arxiv.org/abs/2112.03540
In the first applications lecture of the semester, we build a general framework for generating the real sparks of AGI: linear inverse problems. argmin.net/p/inverse-problem…
Samuel Vaiter retweeted
We organize a Paris meetup for the @LogConference 🤓 Everyone is welcome, whether you can attend the main conference (recommended!) or not 🙃 please submit your proposals!
sites.google.com/view/learni…
We had 2 successful previous editions of the #LoG
@LogConference meetup event in #Paris, with 10+ insightful talks and 15+ amazing poster presentations 😎 Now, we invite you again to submit your proposals to showcase your research in #graph #ML! Form: tinyurl.com/y7fvyb6t
Samuel Vaiter retweeted
🚀 Exciting News! Benchopt 1.6.0 is here with a ton of new features to make ML benchmarking easier than ever! 🎉
🔗 Changelog: benchopt.github.io/whats_new…
🧵👇
Samuel Vaiter retweeted
Want to do research in ML for science in NYC and develop the next generation of foundation models for scientific data? Apply here to be a postdoc at @PolymathicAI @SimonsFdn : forms.gle/JbdWA3VczH3XfgVg7
Samuel Vaiter retweeted
Glad to have been able to host the latest INRIA's benchopt sprint @OwkinScience and thus meet the next generation of ML researchers ! I took this occasion to implement a new FLamby bench github.com/owkin/benchmark_f… now accessible directly from the website: benchopt.github.io/results !
Samuel Vaiter retweeted
We just finished a very productive benchopt sprint, gathering 35 participants over 3 days, hosted at @Owkin_Science, Google AI Hub & SCAI @Sorbonne_Univ.
We merged 11 PRs, updated 9 benchmarks, and developed 7 new ones.
Here are some key highlights!
notes.inria.fr/s/XDBWzL8ID
Samuel Vaiter retweeted
Now availaible in Information and Inference
academic.oup.com/imaiai/arti…
Very proud of this work with @vaiter and @RemiGribonval !
We uploaded our new preprint "A theory of optimal convex regularization for low-dimensional recovery", with @RemiGribonval, @vaiter
hal.archives-ouvertes.fr/hal…
We ask ourselves what is the "best" way to regularize inverse problems given a low dimensional model (e.g. sparse)?
1/3
Samuel Vaiter retweeted
🏜️ New preprint with @vaiter and @BouretReynaud 🏜️
arxiv.org/abs/2405.18828
We propose a Spiking Neural Network which provably learns a classification task by producing neuronal assemblies, whose neurons activity is modeled by Hawkes processes.
1/3
