🔥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

Sebastian Seung has been on a roll today with a very cool thread on some of the history surrounding neural nets and today's prize announcement. Highly recommend reading the whole thing! 👇

Hitchhiker's guide to rebranding: - Machine learning -> statistical mechanics - Loss function -> energy functional - Optimize the model -> minimize free energy - Trained model -> reached equilibrium distribution - KL divergence -> free energy difference - Gaussian noise -> random

🪩The State of AI 2024 has landed! 🪩 Our seventh installment is our biggest and most comprehensive yet, covering everything you *need* to know about research, industry, safety and politics. As ever, here's my director’s cut (+ video tutorial!) 🧵

JAX is fast. When you switch to the JAX backend in Keras 3 (which is XLA compiled by default), you will likely see a 1.5x to 2.5x throughput improvement if coming from torch (eager), or 1.2x to 2x if coming from Keras 2 (i.e. TF graph mode). Sounds too good to be true, so don't

Rethinking Academia: What’s Broken and How We Can Fix It | #2 with Prof. Peter Dayan Prof. Peter Dayan shares his thoughts on some key academic challenges. He highlights the complexities of the academic publishing system. The pressure to publish in high-impact journals creates

Missed #SNUFA24 spiking neural network workshop last week? No worries, we uploaded all 11 talks and the flash talks to Youtube. Check it out, along with talks from every year since we started in 2020 📺 71 videos, 36k views and 3.4k hours watched so far. youtube.com/playlist?list=…

🎉 Great news! Pylustrator, my open-source tool for creating publication-ready Matplotlib plots, works seamlessly with the latest Matplotlib version! 🖌️✨ Curious how it can transform your figures and save your changes as reproducible Python code? Let me show you! 🧵👇

this is paper is kinda wild. turns out that if you simply ask an LLM to straight out predict a timeseries like this: ``` <history> (t1, v1) (t2, v2) (t3, v3) </history> <forecast> (t4, v4) (t5, v5) </forecast> ``` making sure to prepend the prompt like this: ```

Human cortical pyramidal neurons are larger, exhibit more complex branching patterns, and feature distinctly nonlinear biophysical properties compared to rat cortical pyramidal neurons. Do these unique characteristics make human neurons more functionally complex? Would that

Toby Pohlen From 1500 to 1900, the evidence was that heavier-than-air flight was impossible. It was attempted many times, it never worked. On the other hand, zeppelins and balloons empirically scale quite well...

Urban design isn't magic — there are specific reasons why we like some places more than others. So here are 10 ways to make a street more (or less) interesting...

Dendrites doing AI problems, so great to see the increased activity! nature.com/articles/s4146…

Wrote a blogpost on why correlation often is note even approximately related to causation: medium.com/@kording/why-c…

I wrote a quick new post on "Digital Hygiene". Basically there are some no-brainer decisions you can make in your life to dramatically improve the privacy and security of your computing and this post goes over some of them. Blog post link in the reply, but copy pasting below

They are ants solving a geometric problem and it is great in color.

In 1945, six women pulled off a computing miracle. They programmed the world’s first computer—with no manuals, no training. Then, a SINGLE assumption erased them from tech history for decades. The story of how ONE photo nearly deleted computing’s female founders: 🧵

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. 🧵

Eye candy alert. nature made an immersive landing page for MICrONS. nature.com/immersive/d428…

The optimal batch size is 1 (For suitable definitions of "optimal")