Mitigating distribution shift in machine learning-augmented hybrid simulation

DOI10.1137/23M1615425MaRDI QIDQ6998604zbMATH OpenFDO

Authors Jiaxi Zhao, Qianxiao Li

Publication date 11 April 2025

Published in SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1137/23M1615425

zbMATH Keywords

distribution shifterror analysisfluid dynamicsmachine learningregularization

Mathematics Subject Classification ID

Smoothness and regularity of solutions to PDEs (35B65) Finite difference methods for initial value and initial-boundary value problems involving PDEs (65M06) Stability and convergence of numerical methods for initial value and initial-boundary value problems involving PDEs (65M12) Error bounds for initial value and initial-boundary value problems involving PDEs (65M15) Computational learning theory (68Q32) Artificial neural networks and deep learning (68T07) Navier-Stokes equations for incompressible viscous fluids (76D05) Finite difference methods applied to problems in fluid mechanics (76M20) Diffusion (76R50)

Cites work

• A graph convolutional autoencoder approach to model order reduction for parametrized PDEs
• A Multigrid Tutorial, Second Edition
• An overview of projection methods for incompressible flows
• Confidence regions in Wasserstein distributionally robust estimation
• Deep learning
• Failure-Informed Adaptive Sampling for PINNs
• scientific article; zbMATH DE number 671796 (Why is no real title available?)
• scientific article; zbMATH DE number 979834 (Why is no real title available?)
• Kolmogorov widths and low-rank approximations of parametric elliptic PDEs
• Learning Optimal Multigrid Smoothers via Neural Networks
• Mathematical biology. Vol. 2: Spatial models and biomedical applications.
• Numerical methods for Kohn–Sham density functional theory
• Projection Method I: Convergence and Numerical Boundary Layers
• Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
• Reynolds-averaged Navier–Stokes equations with explicit data-driven Reynolds stress closure can be ill-conditioned
• Subgrid modelling for two-dimensional turbulence using neural networks
• Testing the manifold hypothesis
• The Elements of Statistical Learning
• Turbulence, coherent structures, dynamical systems and symmetry.
• Turbulent Flows

This page was built for publication: Mitigating distribution shift in machine learning-augmented hybrid simulation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6998604)