Show HN: Teaching a 4M-param model to read like a child, with RL and no labels

By rybthrow2 · 2026-10-30 · 1 points · 0 comments

https://rjb7731.github.io/blog/2026/learning-from-scratch/

I have been really interested in reinforcement learning and had a go at trying to teach a model through simulation to read and speak through experience. It’s two small networks (~4M parameters total) trained on a MacBook. It passed generated versions of England’s Year 1 Phonics…

Open the full discussion on BetterNews