Eunice ⚡

NVIDIA just trained a 14-billion-parameter AI using evolution, not calculus. Every AI today learns through backpropagation. It computes gradients, adjusts weights, repeats. It works, but it demands precision hardware and enormous GPU clusters. Evolution Strategies offered an alternative. Mutate the model, test it, keep what works. Like biological evolution. The problem was speed. Random mutations on GPUs were painfully slow. EGGROLL fixes this with one trick. It splits huge random matrices into two small ones per mutation. The model mutates, tests, and keeps what works. Hundreds of thousands of mutations run at once. > 100x faster training throughput > 91% speed of pure inference > Pretrains models using only integers > Competitive with backprop on reasoning > Works on non-differentiable systems It pretrained a language model from scratch using zero gradients. It also matched reinforcement learning methods on math reasoning tasks. Everyone kept scaling the calculus to train massive AIs. It turns out, we just needed to evolve.

Happy Sibling’s Day….

I typically ignore all these false claims attacks. But... You can apologize now. I am officially divorced. I won't post any legal docs online, as I respect privacy of my ex-wife, and I appreciate the time we spent together. I am happy to bet $1 billion USD (or any number you choose) that: I am officially divorced (way before today). If you agree to take the bet, we can get lawyers to validate my divorce agreement, which should be dead simple. This bet offer is valid permanently, whenever you feel ready. But if you don't take it within 24hrs, it clearly shows who has been mis-representing to the public. Moving on to better things to do.

If he can produce a divorce agreement signed by both parties as of today, I will immediately issue a public apology. If he cannot, yet claims in the media and in his book that he is already divorced, then that is a clear misrepresentation to the public — yet another example of him making such claims with confidence.