Work

From Flow-Box Invariants to Global Conservation Laws: Machine Learning-Assisted Detection of Singularities and Near-Integrability

Year: 2026

Type: preprint

Abstract: Conservation laws reveal invariant structure in dynamical systems and constrain their evolution to lower-dimensional level sets. Locally, however, per the flow-box theorem, every $d$-dimensional system generically admits $d-1$ functionally independent conservation laws near a regular point even though, generically, only a few such laws persist globally. Data-driven methods typically learn conservation laws on a prescribed region of phase space from which training samples are drawn, leaving their behavior beyond that region unresolved. We formulate this local-to-global continuation problem and develop an {expanding-domain neural deflation} framework that tracks the number of independent conservation laws over nested domains. The method recovers the expected $d-1$ local invariants and detects their successive breakdown as the domain expands, relaxing to the expected global conservation law limit. It distinguishes the local invariants (associated with Frobenius integrability) from persistent global structure and shows that the domain scale over which additional laws survive reflects proximity to Liouville integrability. Pointwise losses can localize isolated obstructions, whereas in more complicated dynamics the breakdown of our loss functions is gradual and spatially diffuse. The framework thus provides a data-driven approach for studying the persistence and breakdown of (so-far-learned) conservation laws across phase space.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.23760

Authors: Adriazola, Jimmie, Kevrekidis, Ioannis G., Kevrekidis, Panayotis G., Zhu, Wei

Institutions: Arizona State University, Johns Hopkins University, University of Massachusetts Amherst, Georgia Institute of Technology

Language: English

Topic: Model Reduction and Neural Networks

Subfield: Statistical and Nonlinear Physics

Field: Physics and Astronomy

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, U.S. Department of Energy, Stanley Medical Research Institute, Science Foundation Ireland, Seoul National University, Okinawa Institute of Science and Technology Graduate University, Air Force Office of Scientific Research

Awards: Collaborative Research: New Directions in Vortex Dynamics: from Theory and Computation to Physical Experiments, CDS&E: Robust Symmetry-Preserving Machine Learning: Theory and Application, G6110646957, Postdoctoral Fellowship: MPS-Ascend: Coherent Control of Nonlinear Schrodinger Dynamics in the Presence of Uncertainty, CAREER: Theoretically Guaranteed Structure-Informed Machine Learning and Data-Driven Structural Inference

OpenAlex ID: W7222971645

Cites: 1

Cited by:

Related to: 10