🔥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

You can now operate robots by just thinking about it. With your brain signals. WOW. This robot system from Stanford has so much sci-fi vibe and wild implications that I don't even know where to start. NOIR decodes the EEG signal from your head into a library of robot skills.

Our collaboration with Google DeepMind has borne fruit! Or rather, a fruit fly 🪰🙂 Simulated with #MuJoCo physics, trained with imitation learning, capable of realistic locomotion, both on the ground and in the air. 1/n biorxiv.org/content/10.110… github.com/TuragaLab/flyb…

#ICLR2024 Excited to share our energy-based model to allow extracting probabilistic representations for concept bottleneck model, allows for better interpretation 👀 and intervention 🔧 for black box artificial neural networks🤖

Empowering Biomedical Discovery with AI Agents 🤖 arxiv.org/abs/2404.02831 Long-standing ambition for biomedical AI is the development of AI systems that can make major scientific discoveries with the potential to be worthy of a Nobel Prize—fulfilling the Nobel Turing Challenge

10 years of Connectomics at Google and a new human cortex publication in Science! It has been a huge honor to pursue brain mapping advances at Google Research with so many amazing colleagues and collaborators. 🙏 youtube.com/watch?v=VSG3_J…

We're sharing progress toward understanding the neural activity of language models. We improved methods for training sparse autoencoders at scale, disentangling GPT-4’s internal representations into 16 million features—which often appear to correspond to understandable concepts.

Nature research paper: Connectome-constrained networks predict neural activity across the fly visual system go.nature.com/3XBCs8K

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”

Given today’s great news from the #NobelPrize2024, I want to share a couple of personal thoughts on Hopfield Networks. This idea had an enormous impact on at least three large disciplines: Statistical Physics, Computer Science and AI, and Neuroscience.

✨🥰 check out our article - and cover 🤩- about Decoding the Brain in Cell cell.com/cell/fulltext/… We review the mathematics, current approaches, and muse about the future… #BCI #neuraldecoding #neuroAI Thanks to my awesome co-authors Adriana Perez Rotondo,

Crosscoders (published today: transformer-circuits.pub/2024/crosscode…) are a new method allowing us to find features shared across different layers in a model, or even across different models. Identifying the same feature when it persists across layers can simplify our understanding of models.

Excited to announce Centaur -- the first foundation model of human cognition. Centaur can predict and simulate human behavior in any experiment expressible in natural language. You can readily download the model from Hugging Face and test it yourself: huggingface.co/marcelbinz/Lla…

Our new AI paper reveals surprising geometric structure in the LLM-learned concepts: 1) They form brain-like "lobes", 2) they form "semantic crystals" much more precise than it first seems, and 3) the concept cloud is more fractal than round:

"Large language models surpass human experts in predicting neuroscience results" w Xiaoliang (Ken) Luo and BrainGPT.org. LLMs integrate a noisy yet interrelated scientific literature to forecast outcomes. nature.com/articles/s4156… 1/8

Join us for the #A3D3 seminar on Monday, 4/14 at 12pm PST, "Accelerating Scientific Discovery in the Brain in the Age of AI," by Dr. Lu Mi Lu Mi at Georgia Tech GaTech CSE! 🔗 Learn more: indico.cern.ch/event/1514382/

How does the brain work? Scientists are closer to the answer with the largest wiring diagram and functional map of a mammalian brain to date. 🧵