Ph.D. student at CNAM on domain adaptation, generative models, and self-supervised learning for remote sensing 🛰
Previously intern @SonyCSL, @Ircam, @Inriagle-bellier.github.io Paris, FranceJoined October 2018
It was mostly a lie. I'm spending most of my social network time in the other place. I'm impressed with how fast the community moved there. My handle is lebellig (same but without the underscore)
It was mostly a lie. I'm spending most of my social network time in the other place. I'm impressed with how fast the community moved there. My handle is lebellig (same but without the underscore)
From Diffusion Models to Schrödinger Bridges
- A shame not to see this NeurIPS keynote live, by the incredible @ArnaudDoucet1
- SB naturally extends flow/ bridge matching and diffusion models
- Particularly useful for data to data
- Links to OT
neurips.cc/virtual/2024/i…
Are geospatial foundation models really impactful?
🚀🚀🌏
Check it in our new pre-print!
Welcome to **PANGAEA: a global and inclusive benchmark for GFMs**
arxiv.org/abs/2412.04204
a short thread 🧵
Do you work on Deep Learning for remote-sensing or Earth Observation (SSL, generative models, foundation models...) and want to move to the other place 🦋?
I've initiated a pack with researchers in the field so we can find each other again! go.bsky.app/EKVwvde
I made a first scan to find people from here in the other place 🦋, maybe some of them got through: don't hesitate to ping me! (my handle is simply lebellig)
The scrolling experience is so much better there...
New notebook! To get to grips with discrete flow matching paper, I've put together a little notebook to test it on MNIST. I kept the article's notations so it's easy to follow alongside the article!
🐍 notebookv1: github.com/gle-bellier/di…
I've been on twitter for a long time and really liked it and still appreciate the relevance of so many research posts on this platform. So I will not quit but start cross-posting on 🎆🛤️🏙️ where you can find me @ lebellig
What are the good practices to set up the number of epochs when using a cosine learning rate scheduler? Set it to a large number + early stopping to determine the needed epochs and then rerun the experiment?
Very nice paper! I like the decomposition between the encoder learning the deterministic components+spatial/modality alignment and the flow matching/diffusion learning the stochastic small-scale physics. (+noise scaling to avoid overfitting on the deterministic dynamics)
Very nice paper! I like the decomposition between the encoder learning the deterministic components+spatial/modality alignment and the flow matching/diffusion learning the stochastic small-scale physics. (+noise scaling to avoid overfitting on the deterministic dynamics)
Does equivariance matter at scale?
Should a model rather learn equi- and invariances from data or should the architecture have the equiv property? This work provides some insights, and SCALING LAWS for each.
P: arxiv.org/abs/2410.23179
New paper out!
We introduce “Generator Matching” (GM), a method to build GenAI models for any data type (incl. multimodal) with any Markov process. GM unifies a range of state-of-the-art models and enables new designs of generative models.
arxiv.org/abs/2410.20587
(1/5)
Maybe we just need to wait for better SSL models to have better FID with diffusion/flow matching models... (alignement model linear probing accuracy vs aligned diffusion/fm model FID) from REPA article arxiv.org/abs/2410.06940)
I had a great time collaborating with @valeriomarsocci and many others on the Pangaea project! You want to know if your earth observation encoder learned useful representations for dense tasks? Use the Pangaea benchmark! 🛰️
I had a great time collaborating with @valeriomarsocci and many others on the Pangaea project! You want to know if your earth observation encoder learned useful representations for dense tasks? Use the Pangaea benchmark! 🛰️
133 Followers 432 FollowingCopernicus Foundation Models for a Thinking Earth 🌍 Official project account.
Funded by Horizon Europe
Call HORIZON-EUSPA-2022-SPACE-02-55
105K Followers 777 FollowingSome projects I was lucky to be part of AlphaGo tuning, AlphaCode, Gato, ReST, r-Gemma, Imagen3, Veo, Genie, MAI. Ex Berkeley, UBC, Oxford Prof, Google DeepMind
254 Followers 311 FollowingThe Raw Researcher | On a mission to make PhD great again | Unsettled, unbound, and relentlessly curious | @uoft | @VectorInst 🚀
647 Followers 1K Followingsenior researcher @MSFTResearch AI for Science, PhD @DeptofPhysics. Opinions my own. Slowly moving to @megstanley.bsky.social
63K Followers 2K FollowingResearch Scientist at Google DeepMind (WaveNet, Imagen, Veo). I tweet about deep learning (research + software), music, generative models (personal account).
495 Followers 564 FollowingMachine learning researcher in the @BAS_News AI Lab. PhD in ML applications to coastal dynamics @CamUniGeography |🛰🌎 🇦🇶🧊|🏳️🌈🥾🏃♂️🌊🏔 🐶 (he/him)
105K Followers 777 FollowingSome projects I was lucky to be part of AlphaGo tuning, AlphaCode, Gato, ReST, r-Gemma, Imagen3, Veo, Genie, MAI. Ex Berkeley, UBC, Oxford Prof, Google DeepMind
192 Followers 156 FollowingPostdoc on Graph ML and biomedical app @EPFL @epfl_lts4 👀
Prior: PhD @Univ_CotedAzur @3IAcotedazur on #OptimalTransport for Graph Representation Learning.
495 Followers 564 FollowingMachine learning researcher in the @BAS_News AI Lab. PhD in ML applications to coastal dynamics @CamUniGeography |🛰🌎 🇦🇶🧊|🏳️🌈🥾🏃♂️🌊🏔 🐶 (he/him)
4K Followers 366 FollowingAssistant Professor in Computer Science, @mcgillu / @Mila_Quebec. Co-Founder and Chair, @ClimateChangeAI. MIT @techreview Innovator Under 35. he/him/his
927 Followers 423 FollowingAssistant research professor, Research software engineer and coach at Johns Hopkins University. I do physics, deep learning, software, and high jump
129 Followers 124 FollowingResearch Intern @heyjasperai | ML PhD Student @Polytechnique - Interested in Monte Carlo methods, sampling, generative modeling and optimal transport
134 Followers 141 Following@GoogleDeepMind. In the past life, research scientist on vegetation information retrieval and SAR remote sensing at JPL/@NASA.
40K Followers 382 FollowingThis account is inactive. Follow @NASAEarth and subscribe to our newsletter for data releases and other Earthdata news: https://t.co/pOtbqt9QpY
89 Followers 444 FollowingPhD student at AstraZeneca and Chalmers University working on generative models for molecules. Computer Science undergrad at Imperial College.
1K Followers 107 FollowingWe are a research team on artificial intelligence for automotive applications working toward assisted and autonomous driving.
1K Followers 607 FollowingProfessor of Data Science for Crop Systems at Forschungszentrum Jülich and University of Bonn
Publications: https://t.co/RNhDLifZI7