Phase space integrity in neural network models of Hamiltonian dynamics: A Lagrangian descriptor approach

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hasmi, Abrari Noor, Hatzikirou, Haralampos, Susanto, Hadi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914436678156288
author Hasmi, Abrari Noor
Hatzikirou, Haralampos
Susanto, Hadi
author_facet Hasmi, Abrari Noor
Hatzikirou, Haralampos
Susanto, Hadi
contents We propose Lagrangian Descriptors (LDs) as a diagnostic framework for evaluating neural network models of Hamiltonian systems beyond conventional trajectory-based metrics. Standard error measures quantify short-term predictive accuracy but provide little insight into global geometric structures such as orbits and separatrices. Existing evaluation tools in dissipative systems are inadequate for Hamiltonian dynamics due to fundamental differences in the systems. By constructing probability density functions weighted by LD values, we embed geometric information into a statistical framework suitable for information-theoretic comparison. We benchmark physically constrained architectures (SympNet, HénonNet, Generalized Hamiltonian Neural Networks) against data-driven Reservoir Computing across two canonical systems. For the Duffing oscillator, all models recover the homoclinic orbit geometry with modest data requirements, though their accuracy near critical structures varies. For the three-mode nonlinear Schrödinger equation, however, clear differences emerge: symplectic architectures preserve energy but distort phase-space topology, while Reservoir Computing, despite lacking explicit physical constraints, reproduces the homoclinic structure with high fidelity. These results demonstrate the value of LD-based diagnostics for assessing not only predictive performance but also the global dynamical integrity of learned Hamiltonian models.
format Preprint
id arxiv_https___arxiv_org_abs_2604_00473
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Phase space integrity in neural network models of Hamiltonian dynamics: A Lagrangian descriptor approach
Hasmi, Abrari Noor
Hatzikirou, Haralampos
Susanto, Hadi
Machine Learning
Dynamical Systems
37M05, 37M25, 37N30, 65P10, 65P40, 68T07
We propose Lagrangian Descriptors (LDs) as a diagnostic framework for evaluating neural network models of Hamiltonian systems beyond conventional trajectory-based metrics. Standard error measures quantify short-term predictive accuracy but provide little insight into global geometric structures such as orbits and separatrices. Existing evaluation tools in dissipative systems are inadequate for Hamiltonian dynamics due to fundamental differences in the systems. By constructing probability density functions weighted by LD values, we embed geometric information into a statistical framework suitable for information-theoretic comparison. We benchmark physically constrained architectures (SympNet, HénonNet, Generalized Hamiltonian Neural Networks) against data-driven Reservoir Computing across two canonical systems. For the Duffing oscillator, all models recover the homoclinic orbit geometry with modest data requirements, though their accuracy near critical structures varies. For the three-mode nonlinear Schrödinger equation, however, clear differences emerge: symplectic architectures preserve energy but distort phase-space topology, while Reservoir Computing, despite lacking explicit physical constraints, reproduces the homoclinic structure with high fidelity. These results demonstrate the value of LD-based diagnostics for assessing not only predictive performance but also the global dynamical integrity of learned Hamiltonian models.
title Phase space integrity in neural network models of Hamiltonian dynamics: A Lagrangian descriptor approach
topic Machine Learning
Dynamical Systems
37M05, 37M25, 37N30, 65P10, 65P40, 68T07
url https://arxiv.org/abs/2604.00473