Teaching a fruit fly brain to play chess with RL in Replit
amasad · x · 2026-09-16
Developer Andrew Blumson is teaching a fly brain (Male CNS, Google's fruit fly connectome model) to play chess via reinforcement learning inside Replit — an unusual demo combining a biological connectome with RL training.
More from Fun
- Google once capped its self-driving Lexus at 25 mph after every employee broke the hands-on-wheel rule — Scobleizer · 2026-09-16
- Blogger mocks Zuck's claim Meta has 'solved AI pacing' with guardrails and liability threats — bindureddy · 2026-09-16
- Ironic twist: the terminal, once inscrutable, is now where I write most fluently — bigblueboo · 2026-09-16
- Creator gets first audience AI check from Pangram — verdict: fully human-written — MatthewChang · 2026-09-16
- Meme: How it feels to play a historical strategy game not made by Paradox — zetalyrae · 2026-09-16
- Censorship Couldn't Kill Literature — the Post-Literate Society Might — julianweisser · 2026-09-16