Read news on neural networks with our app.
Read more in the app
'Like gene editing for neural networks': MIT researchers used GPT-6 Astra to edit Qwen's open weights directly, boosting math performance
Neural networks are clean
Softmax: Why neural networks need non-linearity? life isn't straight-line simple
Why are neural networks and cryptographic ciphers so similar? (2025)
Types and Neural Networks
The Universal Constraint Engine: Neuromorphic Computing Without Neural Networks
TorchLean: Formalizing Neural Networks in Lean
Chrome 146 Now In Beta With WebNN Origin Trial For Neural Networks In The Browser
Hypernetworks: Neural Networks for Hierarchical Data
Batmobile: 10-20x Faster CUDA Kernels for Equivariant Graph Neural Networks
Neural Networks: Zero to Hero
Evolution: Training neural networks with genetic selection achieves 81% on MNIST
OpenAI experiment finds that sparse models could give AI builders the tools to debug neural networks
Researchers isolate memorization from problem-solving in AI neural networks
Bit is all we need: binary normalized neural networks
The lottery ticket hypothesis: why neural networks work
Do we understand how neural networks work?
A robust, open-source framework for Spiking Neural Networks on low-end FPGAs
Neural networks that learn non-linearity without activation functions [pdf]
Computational complexity of neural networks (2022)
