A non-intrusive approach to index-aware learning

Abstract

We present a non-intrusive version of the index-aware learning framework introduced in arXiv:2309.00958. Index-aware learning itself is an approach for learning the time and parameter dependent solutions of differential-algebraic equations (DAEs), in particular those describing electric circuits. A central feature of the approach is that it ensures the learned solutions to fulfill the inherent constraints of the DAE, such as e.g. Kirchhoff's laws in the case of electric circuits. This is achieved by leveraging a decoupling of the DAE into its differential and algebraic parts, with the non-intrusive version of the approach additionally relying on results from arXiv:2604.20475 and arXiv:2107.07755. We illustrate the approach using a filtered buck converter as an example and compare both the intrusive and non-intrusive versions. The code for the example is openly available.

Symmetry-Aware Feature Learning: A Polynomial Separation for Multi-Index Models
Jivan Waber, Vanessa Piccolo, Yatin Dandi +1
arXiv6 Oct 2026PreprintarXiv licenceFull textProbability StatisticsarXiv
Online adaptive non-intrusive model reduction via manifold interpolation and subspace updates: application to FSI convergence acceleration
Azzeddine Tiba, Florian de Vuyst, Iraj Mortazavi
arXiv15 Sep 2026PreprintarXiv licenceFull textNumerical analysis & scientific computingarXiv
Conformal risk control for model-form uncertainty in parametric non-intrusive reduced-order models
Edgar Jaber, Rémy Vallot, Thibault Dairay +1
HAL1 Sep 2026PreprintCC BY 4.0Full textStatisticsHALarXiv
Non-Intrusive Balancing for Data-Driven Model Reduction with Oblique Projections
Alberto Padovan
arXiv10 Oct 2026PreprintCC BY 4.0Full textMathematical physicsarXiv
Learning Orthogonal Multi-Index Models Beyond Small Initialization: Incremental Learning, Competitive Dynamics and Symmetry
Mo Zhou, Weihang Xu, Simon S. Du +1
arXiv9 Sep 2026PreprintCC BY 4.0Full textStatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{forster2026nonintrusive,
  title = {{A non-intrusive approach to index-aware learning}},
  author = {Peter Förster and Idoia Cortes Garcia and Wil Schilders and Sebastian Schöps},
  howpublished = {arXiv},
  year = {2026},
  month = aug,
  doi = {10.48550/arxiv.2605.30955},
  eprint = {2605.30955},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.30955},
}

RIS

TY  - UNPB
TI  - A non-intrusive approach to index-aware learning
AU  - Peter Förster
AU  - Idoia Cortes Garcia
AU  - Wil Schilders
AU  - Sebastian Schöps
PY  - 2026
DA  - 2026/08/31
PB  - arXiv
DO  - 10.48550/arxiv.2605.30955
UR  - https://arxiv.org/abs/2605.30955
AB  - We present a non-intrusive version of the index-aware learning framework introduced in arXiv:2309.00958. Index-aware learning itself is an approach for learning the time and parameter dependent solutions of differential-algebraic equations (DAEs), in particular those describing electric circuits. A central feature of the approach is that it ensures the learned solutions to fulfill the inherent constraints of the DAE, such as e.g. Kirchhoff's laws in the case of electric circuits. This is achieved by leveraging a decoupling of the DAE into its differential and algebraic parts, with the non-intrusive version of the approach additionally relying on results from arXiv:2604.20475 and arXiv:2107.07755. We illustrate the approach using a filtered buck converter as an example and compare both the intrusive and non-intrusive versions. The code for the example is openly available.
ER  -

Abstract

Related

Cite