CosmicAI Seminar: Surrogate Modeling — Bellinger & Ballester

Surrogate models for expensive physics

CosmicAI is the NSF-Simons AI Institute for Cosmic Origins, headquartered at UT Austin, where researchers build machine learning methods for astronomy and cosmology. The first half of its fall seminar series is devoted to AI surrogate models for physical systems — learned approximations that stand in for simulations too costly to run at the scale modern surveys demand, and the question of how far those approximations can be trusted.

This session is presented by Earl Bellinger and Manuel Ballester.

Practical details

• When: Wednesday, September 9, 1:00–2:00 p.m. Central
• Where: UT Austin, hybrid (in person plus online). The specific building and room are not published on the series page — check the CosmicAI events page before travelling, or join online.
• Cost: Free, with no ticketing; the organizers provide Google Calendar and ICS links instead of a registration form.

The talk title had not been announced when this listing was written — the speakers and date are confirmed, the specific topic is not. This is a research-level seminar aimed at people with a machine learning or astrophysics background.