Epistemic Uncertainty-Aware Barlow Twins Reduced Order Modeling for Nonlinear Contact Problems

Epistemic Uncertainty-Aware Barlow Twins Reduced Order Modeling for Nonlinear Contact Problems

This study presents a method for constructing machine learning-based reduced order models (ROMs) that accurately simulate nonlinear contact problems while quantifying epistemic uncertainty. These purely non-intrusive ROMs significantly lower computational costs compared to traditional full order mod...

Full description

Bibliographic Details
Main Authors: Teeratorn Kadeethum, John D. Jakeman, Youngsoo Choi, Nikolaos Bouklas, Hongkyu Yoon
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Access
Subjects:
Reduced order modeling
data-driven modeling
uncertainty quantification
nonlinear contact problems
material science
Online Access:https://ieeexplore.ieee.org/document/10159524/

Similar Items

Similar Items