The imperative of physics-based modeling and inverse theory in computational science

To best learn from data about large-scale complex systems, physics-based models representing the laws of nature must be integrated into the learning process. Inverse theory provides a crucial perspective for addressing the challenges of ill-posedness, uncertainty, nonlinearity and under-sampling.

Saved in:
Bibliographic Details
Published in:Nature Computational Science Vol. 1; no. 3; pp. 166 - 168
Main Authors: Willcox, Karen E., Ghattas, Omar, Heimbach, Patrick
Format: Journal Article
Language:English
Published: United States Nature Publishing Group 01.03.2021
Subjects:
Artificial intelligence
Big Data
Complex systems
Engineering
Geometry
Gravitational waves
Inverse problems
Machine learning
Partial differential equations
Physics
Power
Science
Simulation
Sparsity
ISSN:2662-8457, 2662-8457
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To best learn from data about large-scale complex systems, physics-based models representing the laws of nature must be integrated into the learning process. Inverse theory provides a crucial perspective for addressing the challenges of ill-posedness, uncertainty, nonlinearity and under-sampling.
Bibliography:SourceType-Scholarly Journals-1
ObjectType-Commentary-1
content type line 14
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:2662-8457
2662-8457
DOI:10.1038/s43588-021-00040-z