analiza aiqbench
The paper uses neural networks to estimate the maximal region of attraction of nonlinear dynamical systems, a control-theory problem. Its main contributions are a network architecture that guarantees valid local Lyapunov functions by construction, a universal-approximation proof on compact domains, and an unsupervised, physics-informed training procedure. For ML readers, the useful point is the design principle of encoding stability constraints into the architecture rather than relying on training to enforce them, though the results are validated only on modest-scale numerical experiments.
Learning Maximal Regions of Attraction on Compact Sets using Taylor Lyapunov Networks
arXiv:2408.17246v3 Announce Type: replace-cross Abstract: We consider the problem of learning maximal Lyapunov functions to compute the maximal region of attraction of nonlinear dynamical systems using neural networks. Two obstacles limit existing neural-network approaches: 1) finite-width approximations can fail to preserve the strict Lyapunov inequalities near the equilibrium, where they become arbitrarily weak, and 2) training is necessarily performed on a bounded computational domain for which the classical Lyapunov and Zubov theorems do not directly apply. To address these issues, first we derive Lyapunov and Zubov theorems directly on a prescribed compact computational domain, characterizing the maximal region of attraction within that domain as the solution of a boundary-value PDE. Second, we introduce the Taylor Lyapunov network, an architecture that prescribes the quadratic leading-order term via the system's linearization and represents the higher-order remainder through a structured cubic factorization, guaranteeing that any network in this class is a valid local Lyapunov function by construction. Finally, we prove that Taylor Lyapunov networks are universal approximators of maximal Lyapunov functions on compact domains, and combine these results into an unsupervised, physics-informed training procedure. Numerical experiments on systems of varying dimensions and stability types show that the proposed architecture is efficient at estimating the maximum region of attraction.
Abstract:We consider the problem of learning maximal Lyapunov functions to compute the maximal region of attraction of nonlinear dynamical systems using neural networks. Two obstacles limit existing neural-network approaches: 1) finite-width approximations can fail to preserve the strict Lyapunov inequalities near the equilibrium, where they become arbitrarily weak, and 2) training is necessarily performed on a bounded computational domain for which the classical Lyapunov and Zubov theorems do not directly apply. To address these issues, first we derive Lyapunov and Zubov theorems directly on a prescribed compact computational domain, characterizing the maximal region of attraction within that domain as the solution of a boundary-value PDE. Second, we introduce the Taylor Lyapunov network, an architecture that prescribes the quadratic leading-order term via the system's linearization and represents the higher-order remainder through a structured cubic factorization, guaranteeing that any network in this class is a valid local Lyapunov function by construction. Finally, we prove that Taylor Lyapunov networks are universal approximators of maximal Lyapunov functions on compact domains, and combine these results into an unsupervised, physics-informed training procedure. Numerical experiments on systems of varying dimensions and stability types show that the proposed architecture is efficient at estimating the maximum region of attraction.
Optimization and Control (math.OC); Machine Learning (cs.LG); Systems and Control (eess.SY)
arXiv:2408.17246 [math.OC]
(or arXiv:2408.17246v3 [math.OC] for this version)
https://doi.org/10.48550/arXiv.2408.17246 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 300 cuvinte
Subiecte conexe
Oracle-Efficient and Parameter-Free Agnostic Smoothed Online Learning
arXiv · 8 oct. 2026
Neural Field Tokenizations with Hierarchy and Spatial Locality Priors
arXiv · 9 sept. 2026
Neural Field Tokenizations with Hierarchy and Spatial Locality Priors
arXiv · 9 iun. 2026
OpenHail: An Event-Driven Gymnasium Environment for Electric Ride-Hailing Fleet Control
arXiv · 28 sept. 2026