Designing accurate emulators for scientific processes using calibration-driven deep models

Jayaraman J. Thiagarajan; Bindya Venkatesh; Rushil Anirudh; Peer-Timo Bremer; Jim Gaffney; Gemma Anderson; Brian Spears

doi:10.1038/s41467-020-19448-8

Nature Communications (Nov 2020)

Designing accurate emulators for scientific processes using calibration-driven deep models

Jayaraman J. Thiagarajan,
Bindya Venkatesh,
Rushil Anirudh,
Peer-Timo Bremer,
Jim Gaffney,
Gemma Anderson,
Brian Spears

Affiliations

Jayaraman J. Thiagarajan: Lawrence Livermore National Laboratory, Center for Applied Scientific Computing
Bindya Venkatesh: School of Electrical, Computer and Energy Engineering, Arizona State University
Rushil Anirudh: Lawrence Livermore National Laboratory, Center for Applied Scientific Computing
Peer-Timo Bremer: Lawrence Livermore National Laboratory, Center for Applied Scientific Computing
Jim Gaffney: Lawrence Livermore National Laboratory, Center for Applied Scientific Computing
Gemma Anderson: Lawrence Livermore National Laboratory, Center for Applied Scientific Computing
Brian Spears: Lawrence Livermore National Laboratory, Center for Applied Scientific Computing

DOI: https://doi.org/10.1038/s41467-020-19448-8
Journal volume & issue: Vol. 11, no. 1
pp. 1 – 10

Abstract

Read online

The success of machine learning for scientific discovery normally depends on how well the inherent assumptions match the problem in hand. Here, Thiagarajan et al. alleviate this constraint by allowing the change of optimization criterion in a data-driven approach to emulate complex scientific processes.

Published in Nature Communications

ISSN: 2041-1723 (Online)
Publisher: Nature Portfolio
Country of publisher: United Kingdom
LCC subjects: Science
Website: https://www.nature.com/ncomms/

About the journal