From the lab to the airwaves 🎙️:

Our PI @ninamiolane.bsky.social joined Anita Okorie on Mind Over Matter for a fascinating conversation on geometric AI, brain complexity, and how neuroimaging is helping to understand neurodegenerative disease.

🎧 Listen here: tinyurl.com/mryct7h9

Interested in how geometric intelligence emerges in brains and machines? 🧠➗🤖

I joined Anita Okorie on Mind Over Matter to discuss geometric AI, brain complexity, and how neuroimaging can advance our understanding of neurodegenerative diseases.

🎧 Episode tinyurl.com/mryct7h9

Been waiting for this one😼! A beautiful, digestible guide to non-Euclidean ML — gorgeous visuals, clean tables, and clear explanations of topology, algebra, and geometry. Dive in 👇

The updated version is also available on arxiv at arxiv.org/pdf/2407.09468

Thank you to the excellent team of editors @iopp-mlresearch.bsky.social and our referees for a truly productive peer review process. Highly recommend this up and coming journal for any unsubmitted topical reviews out there!

Made possible with the (equal) dedication of @naturecomputes.bsky.social Johan Mathe @louisacornelis.bsky.social and the contributions of @abbybertics.bsky.social , Domas Buracas, Hansen Lillemark, @christian-shewmake.bsky.social , @fatihdinc.bsky.social Xavier Pennec and @ninamiolane.bsky.social

Our illustrated guide to non-Euclidian ML is finally published!
Check it out for
⭐️ gorgeous figures (with new additions!) on topology, algebra, and geometry in the field
⭐️ broken down tables for easy reading
⭐️ accessible text, additional refs, and more
iopscience.iop.org/article/10.1...

Congratulations!

Looking forward to seeing the research that will come out of your lab 🤩

Many thanks to @mathildepapillon.bsky.social @naturecomputes.bsky.social J. Mathe @louisacornell.bsky.social @abbybertics.bsky.social D. Buracas H. Lillemark @christian-shewmake.bsky.social @fatihdinc.bsky.social X. Pennec for this collaboration !

In the 20th century, non-Euclidean geometry transformed how we model the world with pen and paper.

Now, it is disrupting how we model the world with machines.

Learn more in our illustrated guide to non-Euclidean ML and AI👇

The era of artificial scientific intelligence is here.

As algorithms generate discoveries at scale, what role remains for human scientists? 🤔

Thanks @plosbiology.org for publishing my perspective @ucsantabarbara.bsky.social @ai-ucsb.bsky.social @ucsbece.bsky.social @ucsb-cs.bsky.social 🌟

YouTube video by WiDS Worldwide WiDS Datathon 2025 Winners Panel Discussion

Watch the 2025 winners explain their solution predicting ADHD across sexes🧠:

📺 youtu.be/i6XdxAa9Bak

Thanks to Valerie Elliott Arianna Zuanazzi @amykooz.bsky.social
Maggie Demkin @bowers-wbhi.bsky.social @kaggle.com @childmindinstitute.bsky.social for organizing!

Congrats to all the participants!

Congrats to the winners of the 2025 Global Datathon on Women's Brain Health:

🥇 Manh Nguyen, Thu Nguyen -FPT University
🥈 Baixue Yao, Xiaoyue Zhang -Flagship Labs 84
🥉 Wei Lyu, Zhida Wang @ucsantabarbara.bsky.social (!)

selected from 1000+ participants around the globe! 🌐

@bowers-wbhi.bsky.social

At last Topological Neural Networks are fast🚀

HOPSE builds an encoder for combinatorial complexes, enabling topological deep learning (TDL) w/o the usual computational cost.

A major step forward for TDL!
@martinca.bsky.social @gbg141.bsky.social @marcomonga.bsky.social @levtelyatnikov.bsky.social

Empirically we find that 𝐇𝐎𝐏𝐒𝐄 👀:

⚡Outperforms (speed and performance) GNNs in larger datasets (MANTRA)
🥇Achieves SOTA on Topological Tasks -- predicting Betti numbers 1 & 2 on MANTRA
📈 Is: faster, more accurate, scalable

(5/6)

TDL actually works at scale! And we believe 𝐇𝐎𝐏𝐒𝐄 lays the foundation for broad applications of TDL ✨

📭 Reach out for collaborations

Special thanks to @levtelyatnikov.bsky.social and @gbg1441 and the team @ninamiolane.bsky.social and @Coerulatus :)
(6/6)

Absolutely proud of this work! Huge thanks to @gbg141.bsky.social @ninamiolane.bsky.social @marcomonga.bsky.social — and of course @martinca.bsky.social, who drove the project, learned on the fly, and kept the enthusiasm high at every turn!

From superposition to sparse codes: interpretable representations in neural networks Understanding how information is represented in neural networks is a fundamental challenge in both neuroscience and artificial intelligence. Despite their nonlinear architectures, recent evidence sugg...

🔵 New paper! We explore sparse coding, superposition, and the Linear Representation Hypothesis (LRH) through identifiability theory, compressed sensing, and interpretability. If you’re curious about lifting neural reps out of superposition, this might interest you! 🤓
arxiv.org/abs/2503.01824

Come find me @gsp-workshop.bsky.social at MILA all week! Can’t wait to be back in Montreal connecting with the community🕸️
I’ll be speaking on purely graph-based (!) Topological Deep Learning first thing tomorrow at 9am. Let’s kick off this workshop in style🍩

Thank you Guillermo Bernárdez, @clabat9.bsky.social @ninamiolane.bsky.social
for making this work possible!
🏠 @geometric-intel.bsky.social @ucsb.bsky.social

🍩TopoTune takes any neural network as input and builds the most general TDL model to date, complete with permutation equivariance and unparalleled expressivity.

⚙️ Thanks to its implementation in TopoBench, defining and training these models only requires a few lines of code.

TopoTune is going to ICML 2025!🎉🇨🇦
Curious to try topological deep learning with your custom GNN or your specific dataset? We built this for you! Find out how to get started at geometric-intelligence.github.io/topotune/

YouTube video by Ann S. Bowers WBHI Computational Neuroimaging in Women’s Brain Health

Watch Here! ⬇️⬇️⬇️
youtu.be/dMYeii6LFxI?...

Join us for the #WiDS Panel on Women’s Brain Health!🧠

Why is sex-specific data crucial? From ADHD under-diagnosis to healthcare gaps, we’ll dive in.

📅 March 12 | 8 AM PT
📍 YouTube -link in comment

With @amykooz.bsky.social @ariannazuanazzi.bsky.social E. Rosenthal T. Silk R. Neuhaus & N. Williams

WiDS Datathon 2025 Panel: Women’s Brain Health YouTube video by Women in Data Science Worldwide

Join LIVE: www.youtube.com/live/v_BiTTH...

Our PI @mbeyeler.bsky.social has received the 2024-’25 Harold J. Plous Memorial Award by @ucsb.bsky.social, in recognition of his “outstanding contributions in research, teaching, and service.“

New article by @ucsbengineering.bsky.social:
engineering.ucsb.edu/news/outstan...

How can neural nets extract *interpretable* features from data—& uncover new science?

👉 Discover our mathematical framework tackling this question w/ identifiability theory, compressed sensing, interpretability & geometry!🌐

By @david-klindt.bsky.social @rpatrik96.bsky.social C. O'Neill H Maurer

New preprint!🚀

Decoding neural representations is a challenge in neuroscience & AI.

👉 Learn how identifiability theory, compressed sensing & interpretability research -w/ a dash of geometry- can help!

@david-klindt.bsky.social @rpatrik96.bsky.social C. O'Neill H. Maurer @ninamiolane.bsky.social

At the core:
1️⃣ Identifiability theory
2️⃣ Compressed sensing
3️⃣ Quantitative interpretability

Our goal is a unified model for LRH, superposition, sparse coding, and AutoInterp—backed by theory and practical insights. 🧠🔍
arxiv.org/abs/2503.01824

From superposition to sparse codes: interpretable representations in neural networks Understanding how information is represented in neural networks is a fundamental challenge in both neuroscience and artificial intelligence. Despite their nonlinear architectures, recent evidence sugg...

We’re looking for feedback and especially criticism before sending this off to a journal. If you’re into neural representations, we’d love to hear your thoughts! 📝🔥
arxiv.org/abs/2503.01824