🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

We are delighted to announce our next speakers for the ELLIS RG are fr Stanford University, the authors of the NeurIPS2023 paper (arxiv.org/abs/2306.04251)-FENG CHEN Daniel Kunin Atsushi Yamamura (山村篤志) & Surya Ganguli on 13th Feb '24 @ 5pm CET on Zoom. Save the Date! Link to join the RG👇

Reminder! Happening Tomorrow! ELLIS

Interested in exactly solvable models of learning dynamics and implicit bias? Come check out our "Get Rich Quick" poster at the HiLD Workshop ICML Conference at 10am! With Daniel Kunin, myself, Clémentine Dominé 🍊, FENG CHEN, David Klindt, Andrew Saxe, and Surya Ganguli.

Really cool theory project on feature learning. If you are at the HiLD workshop ICML Conference check it out!

Great job, it was an honor being part of this amazing project! Congrats to the team 💪

Come check out our #NeurIPS2024 spotlight poster on feature learning tomorrow! 📍East Exhibit Hall A-C #2102 📅Thu 12 Dec 4:30 p.m. — 7:30 p.m. PST

Excited to finally share this work w/ Surya Ganguli. Tl;dr: we find the first closed-form analytical theory that replicates the outputs of the very simplest diffusion models, with median pixel wise r^2 values of 90%+. arxiv.org/abs/2412.20292

1/ Our new paper: “Rethinking Fine-Tuning when Scaling Test-Time Compute: Limiting Confidence Improves Mathematical Reasoning” on how to change training to better exploit test-time compute! co-led by Allan Raventós, w/ Nan Cheng, Surya Ganguli & Shaul Druckmann arxiv.org/abs/2502.07154

Wednesday, April 9th at 11AM: TTIC's Young Researcher Seminar Series presents Daniel Kunin (Daniel Kunin) of Stanford Engineering with a talk titled "Learning Mechanics of Neural Networks: Conservation Laws, Implicit Biases, and Feature Learning." Please join us in Room 530, 5th floor.

🚀 Exciting news! Our paper "From Lazy to Rich: Exact Learning Dynamics in Deep Linear Networks" has been accepted at ICLR 2025! arxiv.org/abs/2409.14623 A thread on how relative weight initialization shapes learning dynamics in deep networks. 🧵 (1/9)