avatar
AstroAI
Developing Artificial Intelligence to Solve the Mysteries of the Universe
  • HOME
  • RESEARCH
  • EARTHAI
  • PEOPLE
  • EVENTS
  • LATEST NEWS
  • LUNCH TALKS
  • SUMMER PROGRAM
  • WORKSHOP
  • APPLY
  • CONTACT
Home AstroAI Workshop 2025
Workshop_abstract
Cancel

AstroAI Workshop 2025

Details Invited Speakers Abstracts Register Schedule Venue Accommodations Code of Conduct

Martin Ying

Flowing Through Stellar Model Uncertainties: The Dartmouth Stellar Evolution Emulator

Presenter: Martin Ying

Title: Flowing Through Stellar Model Uncertainties: The Dartmouth Stellar Evolution Emulator

Date/Time: Thursday, July 10th, 2:30 - 2:45 PM

Abstract: Stellar evolution models are fundamental tools in modern astronomy, with diverse applications ranging from exoplanet characterization to cosmological age determination. Despite their critical role, intrinsic uncertainties in these models have historically been overlooked outside of stellar modeling community, mainly due to computational complexity. With rapidly advancing observational capabilities delivering unprecedented data volumes, there is a pressing need for next-generation stellar evolution models capable of comprehensive uncertainty quantification and efficient statistical analyses.

This work presents absolute age determinations for ten Milky Way globular clusters, explicitly incorporating uncertainties arising from 21 stellar evolution parameters. Our analysis reveals that distance and interstellar reddening contribute over 50% of the total age uncertainty. Among stellar physics parameters, metallicity, α-element enhancement, mixing length, and helium diffusion significantly impact the overall error budget.

We introduce the Dartmouth Stellar Evolution Emulator (DSEE), an innovative normalizing flow-based generative model trained on over eight million evolved stellar models. DSEE provides the first continuous, comprehensive database of stellar evolution models spanning extensive parameter spaces across diverse physical regimes. DSEE will significantly reduce computational complexity, enabling efficient, precise absolute stellar parameter estimations, and can be applied to large-scale statistical analyses. Additionally, DSEE improves accessibility to state-of-the-art stellar models for astronomers with limited computational resources.

-->

© 2025 AstroAI. Some rights reserved.

Powered by Jekyll with Chirpy theme.

A new version of content is available.