DeepMind's Kevin Murphy to unveil interactive scientific benchmark modeled on ARC-AGI-3
anshulkundaje · x · 2026-10-17
Kevin Murphy, research scientist at Google DeepMind and author of the Probabilistic Machine Learning textbooks, will speak at ARC Prize Research Summit 2026 in Boston about his "Model Discovery Agent" and building an ARC-AGI-3-like interactive scientific benchmark.
The core idea: let agents run a few experiments in a partially observed stochastic simulator to collect data D, fit a "causal world model" M, then evaluate it by predicting effects of novel (OOD) interventions — p(y | do(x), M, D) — with no need to assess M directly.
More from AGI Musings
- 开发者放话：AI 智能体做真实场景视觉运动控制设计毫无胜算 — AlexTensor · 2026-10-17
- AI professor claims universe is a simulation built to create superintelligence, will be 'turned off' — AIFlow_ML · 2026-10-17
- Pedro Domingos: the key AI training data is what you generate by interacting with it — pmddomingos · 2026-10-17
- Industry Insider: AI Already Cuts IND Filings by Months and Reduces Animal Studies — nlarusstone · 2026-10-17
- ~30% of US STEM PhD dissertations this year contain AI-generated text — docmilanfar · 2026-10-17
- Wishing the RAM shortage makes school Chromebooks unaffordable — AI spillover hits education — RachelVT42 · 2026-10-17