Upcoming Event: Aerospace Engineering and Engineering Mechanics / Oden Institute

Harrington Fellow Symposium: Scientific Machine Learning for Computational Mechanics

,

Abstract

9:00am - 10:00am: Keynote - Recent Advances in Physics-Informed Deep Learning (George Karniadakis, Brown University)

10:15am - 12:15pm: Session 1 (Charley Taylor, UT Austin, Horacia Espinosa, Northwestern University, Johann Guilleminot, Duke University, Steve Sun, Columbia University)

1:30pm - 2:30pm: Session 2 (Nick Vlassis, Rutgers University, Hannah Lu, UT Austin)

3:00pm - 5:00pm: Poster Session 

Biography

Over the past two decades, improvements in computational power, software, and data availability have significantly expanded the role of Artificial Intelligence (AI) in engineering applications. Initially prominent in image processing and informatics, AI methods are now increasingly applied to directly solve ordinary and partial differential equations (ODEs, PDEs) and constitutive equations in computational mechanics. For example, neural networks have been trained to emulate or replace traditional physics solvers and closed-form energy potentials. Other machine learning (ML) techniques, such as Gaussian processes, reduced order models, and automated model discovery, have also been adopted and further developed to work specifically in the context of computational mechanics.

The symposium objective is to bring together the leading experts in scientific machine learning for computational mechanics and synthesize directions for research and education in the United States in this field of science over the next decade and beyond.