EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics-Informed Neural Networks for High-Frequency and Multi-Scale Problems Using Transfer Learning

Abdul Hannan Mustajab, Hao Lyu, Zarghaam Rizvi, Frank Wuttke

Applied Sciences · 11.04.2024 · 14 (8) · 3204

DOI: 10.3390/app14083204 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 11.04.2024. Verfasst von Abdul Hannan Mustajab und 3 weiteren Autor:innen. Die Arbeit wurde bisher 49-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Erasmus Plus Traineeship, Kiel University, Land Schleswig-Holstein within the funding programme Open Access Publikationsfonds.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Physics-Informed Neural Network (PINN) is a data-driven solver for partial and ordinary differential equations (ODEs/PDEs). It provides a unified framework to address both forward and inverse problems. However, the complexity of the objective function often leads to training failures. This issue is particularly prominent when solving high-frequency and multi-scale problems. We proposed using transfer learning to boost the robustness and convergence of training PINN, starting training from low-frequency problems and gradually approaching high-frequency problems through fine-tuning. Through two case studies, we discovered that transfer learning can effectively train PINNs to approximate solutions from low-frequency problems to high-frequency problems without increasing network parameters. Furthermore, it requires fewer data points and less training time. We compare the PINN results using direct differences and L2 relative error showing the advantage of using transfer learning techniques. We describe our training strategy in detail, including optimizer selection, and suggest guidelines for using transfer learning to train neural networks to solve more complex problems.

Förderung

  • Erasmus Plus Traineeship
  • Kiel University
  • Land Schleswig-Holstein within the funding programme Open Access Publikationsfonds

Zitieren

APA

Mustajab, A. H., Lyu, H., Rizvi, Z., & Wuttke, F. (2024). Physics-Informed Neural Networks for High-Frequency and Multi-Scale Problems Using Transfer Learning. Applied Sciences, 14(8), 3204. https://doi.org/10.3390/app14083204

Vancouver

Mustajab AH, Lyu H, Rizvi Z, Wuttke F. Physics-Informed Neural Networks for High-Frequency and Multi-Scale Problems Using Transfer Learning. Applied Sciences. 2024;14(8):3204. doi:10.3390/app14083204

BibTeX
@article{Mustajab2024,
  author = {Mustajab, Abdul Hannan and Lyu, Hao and Rizvi, Zarghaam and Wuttke, Frank},
  title = {Physics-Informed Neural Networks for High-Frequency and Multi-Scale Problems Using Transfer Learning},
  journal = {Applied Sciences},
  year = {2024},
  volume = {14},
  number = {8},
  pages = {3204},
  doi = {10.3390/app14083204}
}
Kostenlos starten →