AI4S Seminar October
Online RIKEN AIP AI for Science seminar: Ken Osato on differentiable cosmological simulations and field-level analysis.
- When
- Wed, October 21, 2026 · 16:00–17:30 JST
- Where
- Online
- Organizer
- RIKEN Center for Advanced Intelligence Project
- Language
- EN
- Source
- Doorkeeper
Summary
The October session of RIKEN AIP's AI for Science (AI4S) seminar series, held online. Ken Osato of Chiba University speaks on "The Differentiable Universe: From Summary Statistics to Field-Level Analysis".
The talk reviews how cosmology has relied on summary statistics such as correlation functions and power spectra to study dark matter and dark energy, and why these miss information in the nonlinear structures formed by gravity. It then introduces differentiable cosmological simulations, which compute how changes in physical parameters shift the predicted matter distribution, and their use in field-level analysis of galaxy survey data.
About the community
A national AI research center whose Doorkeeper group announces public seminars, workshops and symposia on machine learning theory, mathematical science and AI for science, mostly in English and often streamed online.
#ai-for-science#cosmology#differentiable-simulation#research-seminar#riken