🚨Princeton ML Theory Summer School🚨 Aug 6 - 15, 2024 Speakers: * F. Krzakala/L. Zdeborova * S. Rakhlin * B. Lourerio * M. Austern * D. Krotov * J. Altschuler mlschool.princeton.edu Apply by Mar. 1 Sponsors: ** NSF ** Princeton University: ORFE, Princeton Engineering, PACM, CSML,

First day of RIIAA in Quito! I can't overstate how much joy this picture brings me. People have come from all over the world to talk about AI in a Latin American context. Looking forward to this whole week!

One of the best conferences I have attended in a wonderful place—many of the thought leaders of GNNs and GDL in one place. Thanks Pablo Barceló for organizing!

How likely is it that two individuals randomly meet and both happen to be carrying a copy of Olver's Invariant Theory? :) [I am on the left, Ben Blum-Smith on the right. Photo by Soledad Villar]

Join us for today's IAIFI Colloquium at 2:00 pm ET (Friday, March 22) with Soledad Villar (Johns Hopkins University) discussing "Exact and approximate symmetries in machine learning." Watch on YouTube: youtube.com/channel/UCueoF…

Congrats 🎉 to Mathias Unberath Soledad Villar Daniel Khashabi 🕊️ for winning the inaugural "Junior Faculty Awards" from the newly-established Data Science and AI institute at Johns Hopkins University Johns Hopkins Engineering

Encontré muy interesante este artículo de Hogg y Villar sobre la ontología y la epistemología de machine learning, y su adecuación o conflicto con la práctica científica, arxiv.org/abs/2405.18095

LoG Conference 2024 is back !!!👉 We are looking for more reviewers! We have a special emphasis on review quality via monetary rewards, a more focused conference topic, and low reviewer load (max 3 papers). But for this we need your help! Sign up here: forms.gle/Nuff4ndVZDFTis…!

🚀 We’re proposing the 3rd edition of the NeurReps workshop at #NeurIPS2024 and we need your support! 🧠 If you’re a past reviewer or interested in joining the Program Committee, there are many ways to get involved. Please fill out this form: forms.gle/fYeTJcBprjBb5Z…

Super proud of my first PhD student Teresa Huang After her internship at Apple Research NYC this Summer she will join Flatiron CCM as a postdoc this Fall

Thank you Sole for always having the utmost faith in me and the tremendous support throughout my PhD ❤️I am so grateful for all my mentors, collaborators, friends, and family. Hello, New York City 🥳

The Call for Papers for the Learning on Graphs Conference (LoG 2024) is out! Submission deadline: September 11th, 2024 Final decision: November 13th, 2024 LoG is one of the best venues for "all things graph" (Graph ML, GNNs, ML for Molecules, RAG+Graphs, ...) - now in its third

This position paper by David W Hogg and Soledad Villar excellently vocalizes the shift in thinking I noticed when transitioning from astronomy research to ML research — a great read for anyone interested in the intersection of ML and the natural sciences! arxiv.org/abs/2405.18095

Our paper got a prize :) Cheers to lead author Johann Brehmer, and fellow co-authors Sönke Behrends, and Taco Cohen. Our results hint that yes, also at large scale of data and compute, if your data has symmetries, you might be better off building these into your network.

Enjoyed reading this paper! It randomly made me think of the Blow-Up Lemma and the Key Lemma (and later works which had more of an algebraic flavour by Szegedy Balázs). Perhaps nothing more than just family resemblance? arxiv.org/abs/2505.23599 (Soledad Villar)

When does the performance of an ML model transfer across dimensions? arxiv.org/abs/2505.23599 Kudos to my terrific collaborators Eitan Levin, Yuxin Ma, and Soledad Villar. 🧵(1/n)

Starting today I'm embarking upon a new adventure: 12 talks over the next 18 days as part of the London Mathematical Society Hardy Lecture Tour (plus a few additions). Titles and abstracts can all be found here: emilyriehl.github.io/talks/ and slides (when available) will eventually be added.

📢 Call for Papers: NeurReps 2025 ‼️‼️‼️ 🧠 Submit your research on symmetry, geometry, and topology in artificial and biological neural networks. Two tracks: Proceedings (9 pages) and Extended Abstract (4 pages). Deadline: Aug 22, 2025. neurreps.org/call-for-papers