Johan Mathe
@johmathe
Followers
518
Following
75
Media
6
Statuses
96
Builder - applied maths, optimization and dl + chaotic dynamics
San Francisco
Joined May 2009
Very proud of the Atmo team and the recognition from TIME.
time.com
Groundbreaking innovations that are making the world better, smarter, and a little more fun
Atmo is one of the TIME best inventions of 2024!. We are on a mission to make ultra-precise weather forecasting universally accessible. Atmo is honored to be recognized by TIME for our contribution. See the full list here:
2
12
32
RT @alevy: "AGI for clouds" is exactly what we're doing @atmo_ai and @RainmakerCorp. The models are running next token prediction on the at….
0
8
0
RT @mathildepapillo: Our illustrated guide to non-Euclidian ML is finally published!.Check it out for .⭐️ gorgeous figures (with new additi….
0
53
0
RT @SGRodriques: Today, we’re announcing the first major discovery made by our AI Scientist with the lab in the loop: a promising new treat….
0
717
0
We will be hosting our Atmo keynote and social on April 24th. Theme this year is Rapid Intensification :)
2
2
8
RT @tarinziyaee: Remember Moravec's paradox? . "Corner cases". "Long tails". You can play whack-a-mole with them. Or, you can confront t….
0
28
0
RT @mathildepapillo: This should be fun! I'll be sharing our recent literature review on all things geometric/topological/algebraic ML✨🍩.Wi….
0
14
0
RT @johannbrehmer: Does equivariance matter when you have lots of data and compute?. In a new paper with Sönke Behrends, @pimdehaan, and @T….
arxiv.org
Given large datasets and sufficient compute, is it beneficial to design neural architectures for the structure and symmetries of each problem? Or is it more efficient to learn them from data? We...
0
61
0
RT @mathildepapillo: What if any neural network could go topological? While being just as (or more) general, equivariant, and expressive as….
0
52
0
RT @SMataigne: Thrilled to announce that our work about "Bispectrum convolutional layers" for cheap and efficient Group-Invariant NN has be….
0
5
0
If you are attending ICML don’t miss @ninamiolane’s presentation on AI/ML/Topological and Algebraic structures. She is unmatched at making these fairly abstract concepts feel so clear, tangible and useful!. On top of this the visuals (slides, tables) are next level. Again big.
Looking forward to the @GRaM_org_ workshop #ICML2024 !🌟. I'll present a guide to AI/ML with geometric, topological & algebraic structures. W/ @naturecomputes @johmathe @mathildepapillo D. Buracas @hansenlillemark @cashewmake2 @AbbyBertics X. Pennec.
2
5
29
RT @mathildepapillo: Building these structures into ML models makes them more efficient, accurate, and robust💪 We classify and illustrate s….
0
8
0