🔬 An Oscar, Two Asteroids, and the Algorithm in Your sklearn: John Platt on AI for Science
- ID
- 27439
- Status
- summarized
- Published
- 23 Sep 2026, 5:07 AM
- Fetched
- 23 Sep 2026, 6:11 AM
- Provider
- Latent Space
- Category
- developer-ai
- Original URL
- https://www.latent.space/p/john-platt
- Source URL
- https://www.latent.space/feed
Summary
- Score
- 7.0
- Created
- 23 Sep 2026, 6:11 AM
- Tags
- Audience
- developersai_ml_learnersai_agent_users
What happened
Google's John Platt—known for inventing the SMO algorithm in sklearn and winning a technical Oscar—discusses Google's Empirical Research Assistance (ERA), an 'auto-Kaggle' AI system that uses LLMs with Monte Carlo Tree Search to automatically solve scoreable scientific problems. ERA keeps a running tree of experiment notebooks, uses Upper Confidence Bound to pick promising branches, and has Gemini propose ~10 mutations per iteration to maximize a score function.
Why it matters
ERA's architecture—LLM + MCTS over experiment notebooks—is a concrete, reproducible pattern for building autonomous research/optimization agents. If you build AI agents, the 'scoreable task' framing and tree-search-over-code approach is directly applicable to your own agentic pipelines, and the paper and GitHub repo are available to study.
Discussion angle
How does the 'scoreable task + MCTS over notebooks' pattern compare to what we're building today—could this approach replace human-led experiment loops in our own ML or product optimization workflows?