PreprintarXiv licencearXiv
Extrapolating learned dynamics across bifurcations based on generalized synchronization
Abstract
Predicting dynamics across bifurcations by extrapolating trained models is crucial in diverse fields of physics. Generalized synchronization provides a framework for understanding dynamical reconstruction by neural networks, especially in the context of reservoir computing. Building on this framework, we derive a formula combining pretrained readouts to reconstruct unseen chaotic dynamics. Using classical Lagrange interpolation theory, we derive rigorous bounds on readout errors and conjugacy residuals. Numerical examples demonstrate accurate prediction of chaotic dynamics across bifurcations.
Related
- Extrapolating escape dynamics from non-escaping trajectories in Hamiltonian systems
- Generalized Hopf bifurcations: a homological analysis and obstructions to their existence
- Increased-Order Generalized Synchronization in Memristive Neurons: Analytical and Data-Driven Approaches
- Polymerase-mediated quasispecies dynamics: bifurcations, complementation, and robustness of τ-tipping
- Risk-Sensitive Learning in Population Games under Extreme Events: Bifurcations and Chaotic Dynamics
BibTeX
@misc{chen2026extrapolating,
title = {{Extrapolating learned dynamics across bifurcations based on generalized synchronization}},
author = {Lingfeng Chen and Masanobu Inubushi},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.34471},
eprint = {2610.34471},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.34471},
}
RIS
TY - UNPB
TI - Extrapolating learned dynamics across bifurcations based on generalized synchronization
AU - Lingfeng Chen
AU - Masanobu Inubushi
PY - 2026
DA - 2026/10/25
PB - arXiv
DO - 10.48550/arxiv.2610.34471
UR - https://arxiv.org/abs/2610.34471
AB - Predicting dynamics across bifurcations by extrapolating trained models is crucial in diverse fields of physics. Generalized synchronization provides a framework for understanding dynamical reconstruction by neural networks, especially in the context of reservoir computing. Building on this framework, we derive a formula combining pretrained readouts to reconstruct unseen chaotic dynamics. Using classical Lagrange interpolation theory, we derive rigorous bounds on readout errors and conjugacy residuals. Numerical examples demonstrate accurate prediction of chaotic dynamics across bifurcations.
ER -