Beyond white- and black-box modeling tools in optical communications and optical computing: physics-informed data-driven modeling
Abstract
Efficient optimization and control of photonic computing and communication systems increasingly rely on accurate surrogate models/digital twins. While data-driven models may achieve faster inference than traditional physics-based methods, they typically suffer from poor training data efficiency and limited generalizability. To address this trade-off, physics-informed data-driven modeling has emerged as a powerful hybrid paradigm. This paper presents a comparative analysis of these three modeling paradigms across three test use cases: optical amplifiers, directly modulated lasers, and interferometer meshes. By evaluating model complexity, data efficiency, generalizability, and modularity, this work provides a detailed analysis of the respective trade-offs and highlights the advantages of combining physical insight with data-driven learning.
Related
- Mechanism-informed dynamic surrogate modeling for rapid thermal-aberration prediction in bonded optical systems
- Quantification of the habitual axial strains in the murine tibia using a musculoskeletal modeling informed Finite Element analysis approach
- Digital Twin Modeling of Quantum Dynamical Systems: Dissipative Quantum Reservoir Computing
- Continuous vision transformer-based neural operator for scalable beam propagation modeling in optical fibers
- LES-derived turbulence tank optical propagation with numerical adaptive-optics modeling: quantifying OAM modal crosstalk and mitigation
BibTeX
@misc{teofilovic2026beyond,
title = {{Beyond white- and black-box modeling tools in optical communications and optical computing: physics-informed data-driven modeling}},
author = {Isidora Teofilovic and Sergio Hernandez Fernandez and Metodi P. Yankov and Christophe Peucheret and Darko Zibar and Francesco Da Ros},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.07274},
eprint = {2607.07274},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.07274},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Beyond white- and black-box modeling tools in optical communications and optical computing: physics-informed data-driven modeling
AU - Isidora Teofilovic
AU - Sergio Hernandez Fernandez
AU - Metodi P. Yankov
AU - Christophe Peucheret
AU - Darko Zibar
AU - Francesco Da Ros
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2607.07274
UR - https://arxiv.org/abs/2607.07274
AB - Efficient optimization and control of photonic computing and communication systems increasingly rely on accurate surrogate models/digital twins. While data-driven models may achieve faster inference than traditional physics-based methods, they typically suffer from poor training data efficiency and limited generalizability. To address this trade-off, physics-informed data-driven modeling has emerged as a powerful hybrid paradigm. This paper presents a comparative analysis of these three modeling paradigms across three test use cases: optical amplifiers, directly modulated lasers, and interferometer meshes. By evaluating model complexity, data efficiency, generalizability, and modularity, this work provides a detailed analysis of the respective trade-offs and highlights the advantages of combining physical insight with data-driven learning.
ER -