Project 1

Project 1: AI-Accelerated Simulations with Multi-Scale Astrophysics Using Physics-Based Deep Learning

Lead PIs: Vicky Kalogera (NU; vicky at northwestern.edu), Aggelos Katsaggelos (NU; a-katsaggelos at northwestern.edu)
Co-PIs: William Moses (UIUC), Aravindan Vijayaraghavan (NU)
Collaborators: Manuel Ballester Matito (NU), Souvik Chakraborty (Indian Institute of Technology Delhi), Ugur Demir (NU), Seth Gossage (NU), Yongseok Jo (NU), Patrick Koller (NU), Matthew Krafczyk (UIUC), Shamal Lalvani (NU), Santiago López Tapia (NU), Philipp Rajah Moura Srivastava (NU), Tjitske Starkenburg (NU), Christoph Würsch (OST, Switzerland), Hongbo Xia (NU) 
Project Summary: This project has the following objectives: (1) model stellar structure and evolution, (2) combine learned components (trained to emulate simulation) with numeric methods, (3) order-of-magnitude speedups, (4) enable modeling large populations of (single and binary) stars, and (5) incorporate uncertainty quantification. These objectives will be met using (1) physics-informed neural networks (PINNs) and physics-informed neural operators (PINOs), (2) hybrid combination of physics-based deep learning (PBDL) with classical methods, and (3) active learning/mesh refinement.