auto-research
Autonomous research system where agents independently optimize toward research objectives (like minimizing training loss) without human intervention, given objectives, metrics, and constraints
As defined in Skill Issue: Andrej Karpathy on Code Agents, AutoResearch, and the Loopy Era of AI
Where it comes up · 1
Most important claims · 2 of 2Search all claims for “auto-research” →
- Auto-research only works for problems with objective metrics that are easy to evaluate; it fails for subjective optimization goals where you can't automatically verify if a solution is good.Skill Issue: Andrej Karpathy on Code Agents, AutoResearch, and the Loopy Era of AI
- Karpathy ran auto-research on NanoChat hyperparameter tuning overnight and it discovered improvements (weight decay on value embeddings, Adam beta tuning) that he had missed despite two decades of manual hyperparameter optimization experience, suggesting joint interactions between parameters that resist human discovery.Skill Issue: Andrej Karpathy on Code Agents, AutoResearch, and the Loopy Era of AI