The Robot and I: Learning the evolution of gas from molecular clouds to stars

Star formation results from a complex, non-linear interplay between gravity, turbulence, magnetic fields, radiation and stellar feedback.  Machine learning, combined with numerical simulations, can effectively cut through the messiness of observational data to provide insights into the underlying physical behavior.  I will describe several recent studies that apply supervised and unsupervised machine learning methods to tackle fundamental star formation questions: What is the distribution and impact of stellar feedback in molecular clouds? How does dense gas become star-forming? How do star-forming regions evolve?

 

 

Professor Stella Offner is a computational astrophysicist at the University of Texas at Austin who holds appointments in the Astronomy Department and the Oden Institute for Computational Engineering and Sciences.  Dr. Offner's research explores how stars form by combining computer models, telescope observations, and statistical techniques. A core focus of her research aims to develop methods to effectively use artificial intelligence to analyze astronomy data and predict properties that cannot be directly observed. She is the Director of the NSF-Simons AI Institute for Cosmic Origins (CosmicAI) and Co-Director of the Oden Center for Scientific Machine Learning. Dr. Offner completed her PhD in Physics at UC Berkeley and was a Hubble Postdoctoral Fellow at Yale and an NSF Postdoctoral Fellow at the Harvard-Smithsonian Center for Astrophysics. She is the recipient of a variety of awards and honors, including an NSF Early Career Award, a Cottrell Scholar Award, and a Peter O’Donnell Distinguished Researcher Award.