EUVIMEDEUVIMED entdecken
Crossref

Structure- and Asymptotic-Preserving Deep Neural Surrogates for Uncertainty Quantification in Multiscale Kinetic Equations

Wei Chen, Giacomo Dimarco, Lorenzo Pareschi

Multiscale Modeling & Simulation · 30.06.2026 · 24 (2) · 861-888

DOI: 10.1137/25m1768709 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Multiscale Modeling & Simulation, online erschienen am 23.06.2026. Verfasst von Wei Chen und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: China Scholarship Council, National Natural Science Foundation of China, Ministry of University and Research.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The high dimensionality of kinetic equations with stochastic parameters poses major computational challenges for uncertainty quantification (UQ). Traditional Monte Carlo (MC) sampling methods, while widely used, suffer from slow convergence and high variance, which become increasingly severe as the dimensionality of the parameter space grows. To accelerate MC sampling, we adopt a multiscale control variates strategy that leverages low-fidelity solutions from simplified kinetic models to reduce variance. To further improve sampling efficiency and preserve the underlying physics, we introduce surrogate models based on structure- and asymptotic-preserving neural networks (SAPNNs). These deep neural networks are specifically designed to satisfy key physical properties, including positivity, conservation laws, entropy dissipation, and asymptotic limits. By training the SAPNNs on low-fidelity models and enriching them with selected high-fidelity samples from the full Boltzmann equation, our method achieves significant variance reduction while maintaining physical consistency and asymptotic accuracy. The proposed methodology enables efficient large-scale prediction in kinetic UQ and is validated across both homogeneous and nonhomogeneous multiscale regimes. Numerical results demonstrate improved accuracy and computational efficiency compared to standard MC techniques.

Förderung

  • China Scholarship Council
  • National Natural Science Foundation of China
  • Ministry of University and Research
  • Royal Society
  • Future of Artificial Intelligence Foundation
  • Centro Nazionale di Ricera in High Performance Computing, Big Data and Quantum Computing
  • Ministero dell’Universita e della Ricerca
  • NextGenerationEU

Zitieren

APA

Chen, W., Dimarco, G., & Pareschi, L. (2026). Structure- and Asymptotic-Preserving Deep Neural Surrogates for Uncertainty Quantification in Multiscale Kinetic Equations. Multiscale Modeling & Simulation, 24(2), 861-888. https://doi.org/10.1137/25m1768709

Vancouver

Chen W, Dimarco G, Pareschi L. Structure- and Asymptotic-Preserving Deep Neural Surrogates for Uncertainty Quantification in Multiscale Kinetic Equations. Multiscale Modeling & Simulation. 2026;24(2):861-888. doi:10.1137/25m1768709

BibTeX
@article{Chen2026,
  author = {Chen, Wei and Dimarco, Giacomo and Pareschi, Lorenzo},
  title = {Structure- and Asymptotic-Preserving Deep Neural Surrogates for Uncertainty Quantification in Multiscale Kinetic Equations},
  journal = {Multiscale Modeling & Simulation},
  year = {2026},
  volume = {24},
  number = {2},
  pages = {861--888},
  doi = {10.1137/25m1768709}
}
Kostenlos starten →