Andrea Montanari

The risk of AI for education is not students cheating in exams, it is people in general cheating themselves into believing they understand things they don’t.

At NeurIPS next week (giving talks Wed and Sat). Do reach out if you want to chat.

The new call for Motwani postdocs application is now open!
academicjobsonline.org/ajo/jobs/30865

BTW-

Not quite ready for a postdoc? We updated the TCS Masters programs spreadsheet:
www.cs.princeton.edu/~smattw/mast...

Any career stage and in the (SF) Bay Area?
Save the date for TOCA-SV on 11/7!

The Mathematics of Large Machine Learning Models (Lecture 1) by Andrea Montanari YouTube video by International Centre for Theoretical Sciences

Turing Lectures at ICTS
www.youtube.com/watch?v=_fF6...
www.youtube.com/watch?v=mGuK...
www.youtube.com/watch?v=yRDa...

IPAM has been a driving force for the advancement of pure and applied mathematics for a quarter century.

US research and technology has benefited enormously from this and other mathematical centers.

Hello journalists. If you report this you gotta say that under the Constitution he has no power to do it. Thank you for your attention to this matter.

In case you happen to be around :)

Here's a decision that simultaneously captures the Trump administration's policies towards food aid, carbon emissions, and the Biden administration
#food #climate #econsky
marketdesigner.blogspot.com/2025/07/inci...

The 5th Zif summer school on Randomness in Physics and Mathematics has an approaching deadline, May 18 2025! With Benoit Collins, Laszlo Erdös, Laura Foini, Silvio Franz, Tamara Grava, Gaultier Lambert, Andrea Montanari, Ofer Zeitouni.

indico.physik.uni-bielefeld.de/event/220/ov...

Looking forward to visiting again India for this program
www.icts.res.in/program/DSPOM

Shooting ourselves in the other foot.
#econsky #medicine
marketdesigner.blogspot.com/2025/06/visa...

Commencement remarks by President Jonathan Levin The following is the prepared text of remarks for delivery by Stanford President Jonathan Levin at the 134th Commencement Ceremony on June 15, 2025.

Stanford held graduation ceremonies today. Here's Jon Levin's thoughts about the role and value of universities.
#econsky #academicsky
news.stanford.edu/stories/2025...

Nice talk by Pierfrancesco Urbani about our recent work on generalization, overfitting, memorization in two layer networks www.youtube.com/watch?v=kRdH...

One figure summary: learning and overfitting/memorization take place on different timescales in large models.

Opinion | We Are Not Being Asked to Run Into Cannon Fire. We Just Need to Speak Up.

“government of the people, by the people, for the people shall not perish from the earth.”
www.nytimes.com/2025/05/26/o...

Killing the goose that lays the golden eggs (to save on the cost of bird feed, and promote the kinds of viewpoint diversity that the “S” in NSF tries to filter out via peer review…)
marketdesigner.blogspot.com/2025/05/nsf-...

Congratulations to Mark Sellke and Eliran Subag!
www.statslab.cam.ac.uk/rollo-davids...

Dynamical Decoupling of Generalization and Overfitting in Large Two-Layer Networks The inductive bias and generalization properties of large machine learning models are -- to a substantial extent -- a byproduct of the optimization algorithm used for training. Among others, the scale...

arxiv.org/abs/2502.21269

Overfitting is among the conceptually most interesting problems in machine learning.
I am happy of several new phenomena we began to understand with Pierfrancesco Urbani.
Alert: mostly non-rigorous! (Celebrating Jorge Kurchan)
web.stanford.edu/~montanar/OT...

Can't begin to emphasize how insane this is. Could you imagine if, every time a public school teacher needed to get their paycheck, they had to justify to a political appointee how that "aligns with administration priorities"?

NSF halts grant awards while staff do second review Action comes after DOGE team arrives at science agency

BREAKING:The National Science Foundation (NSF) has put a cork in its grantmaking pipeline after billionaire Elon Musk’s Department of Government Efficiency (DOGE) set up shop at the agency this week. | Science | AAAS www.science.org/content/arti...

Spectrum of the Hessian at the empirical risk minimizer for a high--dimensional multinomial regression problem:

Paper:
arxiv.org/abs/2502.01953

Local minima of the empirical risk in high dimension YouTube video by Simons Institute

Nice talk by Basil Saeed at Simons Institute on our work on characterizing the empirical risk landscape via Kac-Rice formula:
www.youtube.com/watch?v=E56W...

Stanford president and provost back Harvard in funding fight The president and provost said Harvard’s rejection of government demands was “rooted in the American tradition of liberty.”

Stanford’s president and provost release a statement supporting Harvard. stanforddaily.com/2025/04/15/l...

Announcing : The 2nd International Summer School on Mathematical Aspects of Data Science
mathsdata2025.github.io
EPFL, Sept 1–5, 2025

Speakers:
Bach @bachfrancis.bsky.social
Bandeira
Mallat
Montanari
Peyré @gabrielpeyre.bsky.social

For PhD students & early-career researchers
Apply before May 15!

In case you are around: sites.google.com/view/jksfest...

Best science book ever?

Top of @nytimes.com app and website right now. Trump critics muzzle themselves because they are afraid of retaliation, including physical threats to their families. www.nytimes.com/2025/03/06/u...

There is no doubt a reason why this quote surfaced in my mind:
“The founding of libraries was like constructing more public granaries, amassing reserves against a spiritual winter which by certain signs, in spite of myself, I see ahead…”

The best short piece I've read on Trump's new Predator America- Brilliant and crucially important @FT editorial

America has turned on its friends - on.ft.com/3ERSO6o via @FT

Protect NIH Research: Advocate for Full Funding As researchers representing universities across the country, the last few weeks have been filled with uncertainty. As you are aware, the Trump administration’s National Institutes of Health (NIH) prop...

A letter of support for the NIH funding of biomedical research, and the damage wrought by imposing severe caps on indirect costs: docs.google.com/forms/d/1Agz...