You are here

Recent Advances in Data-Intensive Physics-Informed Machine Learning for Accelerating Computational Science

Romit Maulik, Argonne National Laboratory
Qi Tang, Los Alamos National Laboratory
Joshua Burby, Los Alamos National Laboratory
Machine learning techniques have recently shown remarkable results for tackling long-standing challenges in computational science. This minisymposium seeks to showcase recent advances in data-driven modeling for dramatically accelerating computational applications with particular emphasis on physics-informed approaches. We are interested in strategies that can leverage large, potentially real-world data sets for learning structure preserving dynamical systems, developing physics-informed closures for advection-dominated problems, and building hybrid PDE and data-driven modeling approaches that significantly accelerate computational workflows for diverse applications.
Confirmed speakers:
Shivam Barwey - Argonne National Laboratory
Stephen Becker - University of Colorado Boulder
David Bortz - University of Colorado Boulder
Andrew Christlieb - Michigan State University
Yingda Cheng - Michigan State University
Emil Constantinescu - Argonne National Laboratory
Cory Hauck - Oak Ridge National Laboratory
Huan Lei - Michigan State University
Daniel Livescu - Los Alamos National Laboratory
Gianmarco Mengaldo - National University of Singapore
Houman Owhadi - Caltech
Daniel Serino - Los Alamos National Laboratory
Xuping Xie - Los Alamos National Laboratory