{"version":1,"type":"story","url":"https://digestai.news/story/john-platt-describes-googles-empirical-research-assistance-tool-for-sc","json":"https://digestai.news/story/john-platt-describes-googles-empirical-research-assistance-tool-for-sc.json","markdown":"https://digestai.news/story/john-platt-describes-googles-empirical-research-assistance-tool-for-sc.md","slug":"john-platt-describes-googles-empirical-research-assistance-tool-for-sc","headline":"John Platt describes Google’s Empirical Research Assistance tool for scientific scoring","summary":"John Platt, a machine‑learning pioneer with an Oscar and two eponymous algorithms, explains Google’s Empirical Research Assistance (ERA). ERA treats any scientific problem that can be expressed as a scoreable task and uses a Gemini LLM to run a Monte‑Carlo‑Tree‑Search‑style loop: the Upper Confidence Bound rule selects promising notebook branches, Gemini proposes about ten mutations per branch, and the system iterates toward higher scores. Platt notes a step change between Gemini 2.0 and 2.5 that turned the approach from non‑functional to “working great.”\n\nThe team has applied ERA to climate‑related challenges, most notably modeling contrail formation. Contrails account for roughly 1 % of human‑induced global warming, and a single gram of aircraft exhaust can seed ten kilograms of ice crystals. After two years of counterfactual modeling, ERA discovered a simple model that captured previously missed confounders. Platt also mentions a $15,000 contrail‑detection Kaggle competition where winners exploited a half‑pixel label error, illustrating how metric gaming can mislead scientific progress.","keyPoints":["ERA combines Gemini LLM with Monte‑Carlo Tree Search to mutate notebook code and maximize defined score functions.","Using ERA, Platt’s team built a contrail climate model after two years of counterfactual work.","Contrails contribute about 1 % of anthropogenic warming; a gram of exhaust can generate ten kilograms of ice crystals."],"whyItMatters":"Automating scientific problem solving could speed discovery in climate research and other domains, showing how LLM‑driven agents can augment expert work.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["Google","Kaggle"],"models":["Gemini"],"people":["John Platt","Dave Bacon","Richard Feynman","Brian Marsden"]},"firstPublishedAt":"2026-09-22T21:07:39Z","updatedAt":"2026-09-22T21:07:39Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"Latent Space","title":"🔬 An Oscar, Two Asteroids, and the Algorithm in Your sklearn: John Platt on AI for Science","url":"https://latent.space/p/john-platt","publishedAt":"2026-09-22T21:07:39Z","type":"newsletter","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"John Platt describes Google’s Empirical Research Assistance tool for scientific scoring\", 22 September 2026, https://digestai.news/story/john-platt-describes-googles-empirical-research-assistance-tool-for-sc","publisher":"Digest AI","title":"John Platt describes Google’s Empirical Research Assistance tool for scientific scoring","datePublished":"2026-09-22T21:07:39Z","url":"https://digestai.news/story/john-platt-describes-googles-empirical-research-assistance-tool-for-sc"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}