Geometric Intelligence Lab

We present graph-based neural surrogates that enable fast and uncertainty-aware emulation of global hybrid-Vlasov simulations via a probabilistic latent-variable model. Data + code are openly available, and compatible with advances happening in data-driven weather prediction!

Are you interested in ML for space/plasma physics? I'll be presenting “Graph-based Neural Space Weather Forecasting” this Saturday at the ML for physics workshop in San Diego @neuripsconf.bsky.social 🚀

📄 Paper: arxiv.org/abs/2509.19605
💻 Code: github.com/fmihpc/space...

#AI4Science #ML4PS

A special thank you to @ucsantabarbara.bsky.social for the thoughtful article on HerBrain – capturing our mission and the importance of understanding how pregnancy shapes the brain 🧠

tinyurl.com/4939bxu5

#TIMEBestInventions #WomensBrainHealth
@ai-ucsb.bsky.social @bowers-wbhi.bsky.social

Huge thanks to S. Kushner L. Pereira
@adelemyers.bsky.social @louisacornelis.bsky.social L. Pritschet Sheila S. D. Martin de Blas H. Grotzinger I. Stelzer L. Chrastil K. Casaletto K. Jordan @susanacarmona.bsky.social @magdamartinezga.bsky.social @emilyjacobs.bsky.social
🧠

Really enjoyed this recent interview on my work with the Dodgers and how the @geometric-intel.bsky.social lab prepared me for it! Thank you for having me UCSB⚾️ Check it out here science.ucsb.edu/news/geometr...

Congrats to Sarah Kushner Luís F. Pereira @adelemyers.bsky.social @louisacornelis.bsky.social Laura Pritschet Sheila S. Daniel Martin de Blas Hannah Grotzinger Ina Stelzer Liz Chrastil Kaitlin Casaletto Kaya Jordan @susanacarmona.bsky.social @magdamartinezga.bsky.social @emilyjacobs.bsky.social 🧠

Thrilled to see HerBrain on #TIME Best Inventions of 2025!

It shows how the anatomy of the brain changes week-by-week through pregnancy and postpartum 🧠

@geometric-intel.bsky.social @ai-ucsb.bsky.social @bowers-wbhi.bsky.social @neuromaternal.bsky.social !

Rumor has it Mother Nature might be the best AI engineer🌱💡

Come see why at Brass Bear Brewing, Nov 13 @ 6pm. With Dr. Fatih Dinc, we’ll dive into how brains, bots, and biology compute with intelligence

Huge congrats to Simon Mataigne and P.-A. Absil on two new papers out this month! 🎉

Bringing elegant math to life through algorithms on the Stiefel manifold and Riemannian barycenters🌐

The second paper called "On the Approximation of the Riemannian Barycenter" uses the bounds from the above article to compute approximations of the Riemannian barycenter at very low computational cost!

link.springer.com/chapter/10.1...

The first paper called "Bounds on the geodesic distances on the Stiefel manifold for a family of Riemannian metrics" studies the geometry of the Stiefel manifold to derive simple and insightful bounds on the computationally untractable geodesic distance.

sciencedirect.com/science/arti...

Details on our research and how to apply to our lab are on our website: gi.ece.ucsb.xn--edu-v113b
@uofcalifornia.bsky.social @ucsantabarbara.bsky.social
@ucsbece.bsky.social @ucsb-cs.bsky.social

And yes, we do play volleyball on the beach nearby🏐🌴

🚀We’re recruiting PhD students for 2026!
If you’re passionate about uncovering the geometric signatures of intelligence in brains🧠 and AI🤖, we’d love to hear from you!

Join us in exploring the computations that power natural and artificial minds. 🌐✨

The Geometric Intelligence Lab @ UC Santa Barbara The mission of the Geometric Intelligence Lab is to reveal the geometric signatures of natural and artificial intelligence.

Details on our research and how to apply to our lab are on our website: gi.ece.ucsb.edu.

We hope to hear from you! 👋
@uofcalifornia.bsky.social @ucsantabarbara.bsky.social @ucsbece.bsky.social @ucsb-cs.bsky.social @geometric-intel.bsky.social

From your desk you'll see dolphins, beach volleyball courts, surf spots & more importantly: your colleagues!

@adelemyers.bsky.social @mathildepapillon.bsky.social @franciscoacosta.bsky.social @louisacornelis.bsky.social @abbybertics.bsky.social @bariskurtkaya.bsky.social and gi.ece.ucsb.edu/people

AI Curriculum — REAL AI

You are specifically interested in AI?

👉 Look at the AI Curriculum from the UCSB Real AI Initiative offered within the CS and the ECE departments.

ai.ece.ucsb.edu/ai-ucsb

@ai-ucsb.bsky.social @ucsbece.bsky.social @ucsb-cs.bsky.social @ucsbengineering.bsky.social

We are physicists, neuroscientists, mathematicians and computer scientists who study intelligence in biological and artificial neural networks and use our findings to build better AI models.

👉 You can join us via UCSB ECE, CS, Math or DYNS PhD programs!

Two more Nobel Prizes at @ucsantabarbara.bsky.social , congratulations !!

The UC sets a new world record with 5 Nobel Prizes this year!

We can't wait to see your innovative solutions! Full details on the tasks and data are on our site:

🔗 geometric-intelligence.github.io/topobench/td...

#TDLChallenge #AI #ML #TAGinDS

A huge thank you to our partners! 🤝

💰 Cash prizes are sponsored by the brilliant team at Arlequin AI.

🏛️ Internships are hosted by the world-class UC Santa Barbara @geometric-intel.bsky.social and EPFL IMOS labs.

Prizes, research gigs, and a chance to twist data into new shapes with topology + ML? 👾 Count me in for the Topological Deep Learning Challenge 2025! 🌀 Who else is ready to expand the data universe?

Diving into topological deep learning🌀? Check out TopoBench — a unified framework for rigorous benchmarking of TDL models.

Pleased to share it has been accepted to DMLR!

Code & paper: geometric-intelligence.github.io/TopoBench/

Struggling with reproducible benchmarks in TDL? We’ve got you covered.

12+ models tested on 22+ diverse, cross-domain datasets. Tutorials to get you started with topology & higher-order homophily Webpage/Docs: geometric-intelligence.github.io/TopoBench/

Structural Liftings TopoBench is a Python library designed to standardize benchmarking and accelerate research in Topological Deep Learning - geometric-intelligence/TopoBench

Curious about topological liftings? TopoBench lets you experiment with them on your own data! We’ve included 20+ liftings from last year’s TDL Challenge @TAGinDS, @PyT_Team_. Explore them all in our Wiki: github.com/geometric-in... #TDA

At its core, TopoBench is a framework for robust and fair evaluation in TDL. We’re empowering researchers and practitioners to compare models with confidence and push the field forward, together.

For a deep dive into our framework for reproducible benchmarking, our methodology, and key findings, check out the full paper. Paper: arxiv.org/abs/246.06642 #MachineLearning #GNN #DataScience

This was a massive team effort!
Developed & led by @levtelyatnikov.bsky.social , @gbg141.bsky.social , Theodore Papamarkou. With major contributions from @Marco Montagna, Mustafa Hajij, Ghada Zamzmi, Michael T. Schaub, @ninamiolane.bsky.social , & @sscardapane.bsky.social , and many others!

We can’t wait to see what the community builds with TopoBench! Star our repo, read the paper, and bring reproducibility to your TDL research. Contributions and feedback are welcome! #OpenSource #Python #AcademicTwitter