Yacine: RL can run in real time — almost nobody is seriously trying
yacineMTB · x · 2026-09-11
Ex-Tesla engineer Yacine argues that reinforcement learning is far from its speed limit and could run in real time, yet almost no one in the field is seriously pushing that direction.
Related event: Ex-Tesla Engineer: Real-Time RL Is Possible, Nobody's Trying(2 posts)→
More from Research
- Michael Levin's Controversial 'Platonic Space' Paper Clears Peer Review — ZeroStateReflex · 2026-09-11
- Training a 6-Expert MoE GPT-2 From Scratch on a Single RTX 3090 in 8 Days — rasbt · 2026-09-11
- Mathematicians react to the Navier-Stokes proof claim — silver__tsuki · 2026-09-11
- Multi-LLM social worlds are 'large': models may learn heuristics instead of grader-modeling — 1a3orn · 2026-09-11
- Lila Sciences' open-endedness team: 'don't throw out the weird result' drives AI-driven discovery — tw_killian · 2026-09-11
- Researcher 'Turns the Fly Bisexual' Using a 166,606-Neuron Connectome Model — dejavucoder · 2026-09-11