🔥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

1/n I’m very excited to present this Spotlight. It was one of the more creative projects of my PhD, and also the last one with Blake Bordelon ☕️🧪👨‍💻 & Cengiz Pehlevan, the best coauthors you can have :) Come by this afternoon to learn "How Feature Learning Can Improve Neural Scaling Laws."

Adam deserves the award, but in Singapore everyone still uses SGD

It's that time of the year again... Applications are open for students and teaching assistants for Introduction to Computational Neuroscience 2025 summer school. Applications are open till May 15th. Apply now and RT! #ArabCompNeuro25 #neuroscience #education

Nice article! I appreciate that it mentions my work and the work of my students. I want to add to it. It is true that there is some inspiration from spin glasses, but Hopfield is much bigger than spin glasses. The key ideas that resurrected artificial neural networks in 1982

Not unplausible... xkcd.com/3085/ Am looking for Sabine Hossenfelder to weight in

Alex Nichol it represents the forward and backward pass

Elegant theoretical derivations are exclusive to physics. Right?? Wrong! In a new preprint, we: ✅"Derive" a spiking recurrent network from variational principles ✅Show it does amazing things like out-of-distribution generalization 👉[1/n]🧵 w/ co-lead Dekel Galor & Jake Yates

A minimax entropy model accurately predicts large-scale neural activity in the data from over 1,000 mouse #neurons, demonstrating that a large amount of information can be compressed into a small number of correlations. Check it out: go.aps.org/43jOIfI

1/3 Geoffrey Hinton once said that the future depends on some graduate student being suspicious of everything he says (via Lex Fridman). He also said was that it was impossible to find biologically plausible approaches to backprop that scale well: radical.vc/geoffrey-hinto….

Data Mixing Can Induce Phase Transitions in Knowledge Acquisition 𝑨 𝑪𝑳𝑬𝑨𝑵, 𝑭𝑶𝑹𝑴𝑨𝑳 𝑩𝑹𝑬𝑨𝑲𝑫𝑶𝑾𝑵 𝑶𝑭 𝑾𝑯𝒀 𝒀𝑶𝑼𝑹 𝟕𝑩 𝑳𝑳𝑴 𝑳𝑬𝑨𝑹𝑵𝑺 𝑵𝑶𝑻𝑯𝑰𝑵𝑮 𝑭𝑹𝑶𝑴 𝑯𝑰𝑮𝑯-𝑸𝑼𝑨𝑳𝑰𝑻𝒀 𝑫𝑨𝑻𝑨 This paper reveals phase transitions in factual memorization

1. אהוד ברק עשה את כל חייו למען ביטחון מדינת ישראל. האיש שהיה ראש אמ''ן, רמטכ''ל, שר ביטחון וראש ממשלה (וגם הקצין המעוטר ביותר אי פעם), חי ונושם ביטחון. אהוד שניאורסון עשה את כל חייו למען קהיליית המודיעין. הוא צמח ב-8200, זכה בפרס ביטחון ישראל וציון לשבח מהרמטכ''ל והגיע לפקד על

You know all those arguments that LLMs think like humans? Turns out it's not true. 🧠 In our paper "From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning" we test it by checking if LLMs form concepts the same way humans do Yann LeCun Chen Shani Dan Jurafsky

[2505.22749] Self-orthogonalizing attractor neural networks emerging from the free energy principle arxiv.org/abs/2505.22749

How does in-context learning emerge in attention models during gradient descent training? Sharing our new Spotlight paper ICML Conference: Training Dynamics of In-Context Learning in Linear Attention arxiv.org/abs/2501.16265 Led by Yedi Zhang with Aaditya Singh and Peter Latham

מי שהולך לישון עם כלביא, מתעורר עם פשפשיא

Incredibly excited for the UK Neural Computation Conference 2025 @ Imperial College London, 9th - 11th July (main event 10th - 11th) 🇬🇧. World leading scientists working in brain computation - from advanced experimental neuroscience to AI and mathematical modelling (and all

This is the syllabus of the course Geoffrey Hinton and I taught in 1998 at the Gatsby Unit (just after it was founded). Notice anything?

מוזמנות.ים לקרוא על אוטודידקטיות, על סקרנות בלתי נלאית להבנת המוח האנושי, על התגובות מהקהילה המדעית (כולל של פרופ' אילנה גוזס ופרופ' בועז ברק Boaz Barak), ובעיקר על הרצון להפעיל את המוח האישי של אדם אחד - רגע לפני שלא יכל עוד. קישור לכתבה ב'הארץ': haaretz.co.il/magazine/2025-…

On set today with Inbal Rabin-Lieberman, the 26-year-old heroine whose quick thinking and IDF weapons training meant that her village, Nir Am, was the only one along the Gaza border where none of the residents was killed. She’s watching today as they film a reenactment of her

Pleased to say that our story of how a theory of self-supervised learning in cortical layers accounts for several experimental observations is now out 🎉 nature.com/articles/s4146…