Transferable reconstruction of nonlinear network dynamics from sentinel nodes
Abstract
Reconstructing the state of a large networked system from measurements at only a few nodes is a challenge for monitoring, prediction, and intervention. Here we ask whether such reconstruction is possible and whether it can transfer across different nonlinear dynamics. We study this question across various networks and across $16$ nonlinear dynamics on networks from different domains. For each network, we observe only a vanishingly small fraction of sentinel nodes and train either a neural network or linear decoder. We find that accurate reconstruction is often possible, and that transferability is structured rather than universal. Eleven dynamics with adjacency-matrix-type coupling form a robust transferable class. By contrast, five diffusively coupled dynamics form isolated transfer components. Non-random sentinel selection is consistently important, and linear decoders often approach neural-network performance. These results show that sparse node observations can encode enough information to reconstruct full network equilibria across broad families of nonlinear dynamics, while also revealing sharp limits to universal transfer.
Related
- Transferable Graph Neural Network Surrogates for Molecular Dynamics Across Crystal Symmetries
- Quantization-Aware Nonlinear Diffractive Neural Network for Robust Optical Image Reconstruction under Fabrication Constraints
- Unraveling Nonlinear Wave Dynamics in Hydraulic Engineering: Symbolic Neural Network Solutions for the (31)-Dimensional P-Type Model+
- Fading Memory and Nonlinear Dynamics in a Microring Resonator Network and their Use in Computation
- Learning Transferable Self-Supervised Priors for Super-Resolution Reconstruction in Structured Illumination Microscopy
BibTeX
@misc{masuda2026transferable,
title = {{Transferable reconstruction of nonlinear network dynamics from sentinel nodes}},
author = {Naoki Masuda and Bisna Mary Eldo and Tharusha Bandara},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.27968},
eprint = {2609.27968},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.27968},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Transferable reconstruction of nonlinear network dynamics from sentinel nodes
AU - Naoki Masuda
AU - Bisna Mary Eldo
AU - Tharusha Bandara
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2609.27968
UR - https://arxiv.org/abs/2609.27968
AB - Reconstructing the state of a large networked system from measurements at only a few nodes is a challenge for monitoring, prediction, and intervention. Here we ask whether such reconstruction is possible and whether it can transfer across different nonlinear dynamics. We study this question across various networks and across $16$ nonlinear dynamics on networks from different domains. For each network, we observe only a vanishingly small fraction of sentinel nodes and train either a neural network or linear decoder. We find that accurate reconstruction is often possible, and that transferability is structured rather than universal. Eleven dynamics with adjacency-matrix-type coupling form a robust transferable class. By contrast, five diffusively coupled dynamics form isolated transfer components. Non-random sentinel selection is consistently important, and linear decoders often approach neural-network performance. These results show that sparse node observations can encode enough information to reconstruct full network equilibria across broad families of nonlinear dynamics, while also revealing sharp limits to universal transfer.
ER -