A Progressive Instructional Paradigm for Physics-informed Neural Networks
IFS/ACM Transactions on Machine Learning · 13.11.2026 · 3 (1) · 38-47
DOI: 10.70891/tml.2026.040053 ↗Kurz gefasst
Fachartikel in der Zeitschrift IFS/ACM Transactions on Machine Learning, online erschienen am 13.11.2026. Verfasst von Wen Zhang und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Physics-informed neural networks as an interdisciplinary field combining deep learning and physical modeling, which needs high demands on student ability to transfer knowledge across disciplines. To address the cognitive gap students face when transitioning from traditional mathematical modeling to deep learning methods, this paper develops a three-stage progressive instructional paradigm. Taking the one-dimensional heat conduction equation as an example, teacher guides students through three successive cognitive stage, i.e., mechanistic modeling, data-driven modeling, and physics-informed modeling, corresponding respectively to the finite difference method, pure data-driven MLP fitting, and the PINNs approach. Rather than introducing PINNs directly as a new algorithm, the proposed design organizes their learning as a problem-driven cognitive progression. This can help students understand the paradigm shift from traditional numerical computation to data-driven learning and ultimately to physics-informed integration.
Zitieren
APA
Zhang, W., Liu, D., Keerio, I. K., & Li, H. (2026). A Progressive Instructional Paradigm for Physics-informed Neural Networks. IFS/ACM Transactions on Machine Learning, 3(1), 38-47. https://doi.org/10.70891/tml.2026.040053
Vancouver
Zhang W, Liu D, Keerio IK, Li H. A Progressive Instructional Paradigm for Physics-informed Neural Networks. IFS/ACM Transactions on Machine Learning. 2026;3(1):38-47. doi:10.70891/tml.2026.040053
BibTeX
@article{Zhang2026,
author = {Zhang, Wen and Liu, Dongshuai and Keerio, Imran Khan and Li, Haifeng},
title = {A Progressive Instructional Paradigm for Physics-informed Neural Networks},
journal = {IFS/ACM Transactions on Machine Learning},
year = {2026},
volume = {3},
number = {1},
pages = {38--47},
doi = {10.70891/tml.2026.040053}
}
EUVIMED