Deep Adaptive Sampling for Surrogate Modeling Without Labeled Data
Journal of Scientific Computing · 2024-12 · 101 (3)
DOI: 10.1007/s10915-024-02711-1 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Scientific Computing, online erschienen am 08.11.2024. Verfasst von Xili Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 83 Literaturangaben. Gefördert von: Natural Science Foundation of Hunan Province, ShanghaiTech University, National Science Foundation.
Abstract (Originalsprache)
Surrogate modeling is of great practical significance for parametric differential equation systems. In contrast to classical numerical methods, using physics-informed deep learning-based methods to construct simulators for such systems is a promising direction due to its potential to handle high dimensionality, which requires minimizing a loss over a training set of random samples. However, the random samples introduce statistical errors, which may become the dominant errors for the approximation of low-regularity and high-dimensional problems. In this work, we present a deep adaptive sampling method for surrogate modeling of low-regularity parametric differential equations and illustrate the necessity of adaptive sampling for constructing surrogate models. In the parametric setting, the residual loss function can be regarded as an unnormalized probability density function (PDF) of the spatial and parametric variables. In contrast to the non-parametric setting, factorized joint density models can be employed to alleviate the difficulties induced by the parametric space. The PDF is approximated by a deep generative model, from which new samples are generated and added to the training set. Since the new samples match the residual-induced distribution, the refined training set can further reduce the statistical error in the current approximate solution through variance reduction. We demonstrate the effectiveness of the proposed method with a series of numerical experiments, including the physics-informed operator learning problem, the parametric optimal control problem with geometrical parametrization, and the parametric lid-driven 2D cavity flow problem with a continuous range of Reynolds numbers from 100 to 3200.
Förderung
- Natural Science Foundation of Hunan Province
- ShanghaiTech University
- National Science Foundation
- National Natural Science Foundation of China
Zitieren
APA
Wang, X., Tang, K., Zhai, J., Wan, X., & Yang, C. (2024). Deep Adaptive Sampling for Surrogate Modeling Without Labeled Data. Journal of Scientific Computing, 101(3). https://doi.org/10.1007/s10915-024-02711-1
Vancouver
Wang X, Tang K, Zhai J, Wan X, Yang C. Deep Adaptive Sampling for Surrogate Modeling Without Labeled Data. Journal of Scientific Computing. 2024;101(3). doi:10.1007/s10915-024-02711-1
BibTeX
@article{Wang2024,
author = {Wang, Xili and Tang, Kejun and Zhai, Jiayu and Wan, Xiaoliang and Yang, Chao},
title = {Deep Adaptive Sampling for Surrogate Modeling Without Labeled Data},
journal = {Journal of Scientific Computing},
year = {2024},
volume = {101},
number = {3},
doi = {10.1007/s10915-024-02711-1}
}
EUVIMED