EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics-informed time series analysis with Kolmogorov-Arnold networks under Ehrenfest constraints

Abhijit Sen, Illya V. Lukin, Kurt Jacobs, Lev Kaplan, Andrii G. Sotnikov, Denys I. Bondar

Physical Review Research · 06.04.2026 · 8 (2)

DOI: 10.1103/84vm-tvzm ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physical Review Research, online erschienen am 06.04.2026. Verfasst von Abhijit Sen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Army Research Office.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The prediction of quantum dynamical responses lies at the heart of modern physics. Yet, modeling these time-dependent behaviors remains a formidable challenge because quantum systems evolve in high-dimensional Hilbert spaces, often rendering traditional numerical methods computationally prohibitive. While large language models have achieved remarkable success in sequential prediction, quantum dynamics presents a fundamentally different challenge: forecasting the entire temporal evolution of quantum systems rather than merely the next element in a sequence. Existing neural architectures such as recurrent and convolutional networks often require vast training datasets and suffer from spurious oscillations that compromise physical interpretability. In this work, we introduce an approach: Kolmogorov-Arnold networks (KANs) augmented with physics-informed loss functions that enforce the Ehrenfest theorems. Our method achieves superior accuracy with significantly less training data: It requires only 5.4 % of the samples (200) compared to temporal convolution networks (3700). We further introduce the chain of KANs, an architecture that embeds temporal causality directly into the model design, making it particularly well suited for time series modeling. Our results demonstrate that physics-informed KANs offer a compelling advantage over conventional black-box models, maintaining both mathematical rigor and physical consistency while dramatically reducing data requirements.

Förderung

  • Army Research Office

Zitieren

APA

Sen, A., Lukin, I. V., Jacobs, K., Kaplan, L., Sotnikov, A. G., & Bondar, D. I. (2026). Physics-informed time series analysis with Kolmogorov-Arnold networks under Ehrenfest constraints. Physical Review Research, 8(2). https://doi.org/10.1103/84vm-tvzm

Vancouver

Sen A, Lukin IV, Jacobs K, Kaplan L, Sotnikov AG, Bondar DI. Physics-informed time series analysis with Kolmogorov-Arnold networks under Ehrenfest constraints. Physical Review Research. 2026;8(2). doi:10.1103/84vm-tvzm

BibTeX
@article{Sen2026,
  author = {Sen, Abhijit and Lukin, Illya V. and Jacobs, Kurt and Kaplan, Lev and Sotnikov, Andrii G. and Bondar, Denys I.},
  title = {Physics-informed time series analysis with Kolmogorov-Arnold networks under Ehrenfest constraints},
  journal = {Physical Review Research},
  year = {2026},
  volume = {8},
  number = {2},
  doi = {10.1103/84vm-tvzm}
}
Kostenlos starten →