███████╗ █████╗ ██████╗ ██████╗ ██╗ ██╗ █████╗ ████████╗ ██╔════╝██╔══██╗██╔══██╗██╔═══██╗██║ ██║██╔══██╗╚══██╔══╝ ███████╗███████║██████╔╝██║ ██║███████║███████║ ██║ ╚════██║██╔══██║██╔══██╗██║ ██║██╔══██║██╔══██║ ██║ ███████║██║ ██║██████╔╝╚██████╔╝██║ ██║██║ ██║ ██║ ╚══════╝╚═╝ ╚═╝╚═════╝ ╚═════╝ ╚═╝ ╚═╝╚═╝ ╚═╝ ╚═╝
> Sabohat — a brain-inspired AI model based on SNN. The goal — make the model learn on its own and significantly reduce the CL problem (catastrophic forgetting).
In v35 I removed loss.backward() entirely — and unexpectedly the model outperformed BP. Here is how LED (Lateral Error Diffusion) came about, why it worked, and where its limit was.
A film, a name, and the idea of building an AI that works like a brain. This is the story of how the Sabohat project began — why GNN, why biological mechanisms, and why the name changed.