A global structure-preserving kernel method for the learning of Poisson systems
convergence ratesHamiltonian systemsinverse problemsmanifold learningPoisson manifoldsstructure-preserving kernel regression
Relations of finite-dimensional Hamiltonian and Lagrangian systems with topology, geometry and differential geometry (symplectic geometry, Poisson geometry, etc.) (37J39) Discretization methods and integrators (symplectic, variational, geometric, etc.) for dynamical systems (37M15) Symplectic manifolds (general theory) (53D05) Poisson manifolds; Poisson groupoids and algebroids (53D17) Differential geometric methods (tensors, connections, symplectic, Poisson, contact, Riemannian, nonholonomic, etc.) for problems in mechanics (70G45)
In this paper, a novel kernel ridge regression method is proposed that allows the recovery of globally defined nonlinear Hamiltonian functions on Poisson manifolds out of datasets made of noisy observations of Hamiltonian vector fields. The kernel method exhibits various key advantages. The differential equation constraints necessary for structure preservation are incorporated in the kernel ridge regression framework due to the the differential reproducing property. For kernel regression, ill-posedness is inherent in recovering the Hamiltonian function in the presence of Poisson degeneracy due to the presence of Casimir functions. The Tikhonov regularized kernel regression ensures a unique solution due to strict convexity. An operator-theoretic framework and a kernel estimator representation is presented that enables a rigorous analysis of estimation and approximation errors. The estimator converges to a true Hamiltonian function with high probability, as the number of random samples increases. The structure-preserving kernel ridge regression is implemented in Python and extensive numerical experiments are conducted on Hamiltonian systems on symplectic manifolds: the rigid body underwater vehicle dynamics, and recovery of two-vortex dynamics.
- An introduction to manifolds
- Argumentwise invariant kernels for the approximation of invariant functions
- Asymptotic and Lyapunov stability of constrained and Poisson equilibria
- Bifurcation of relative equilibria in mechanical systems with symmetry
- Deep Hamiltonian neural networks based on symplectic integrators
- Derivative reproducing properties for kernel methods in learning theory
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Discrete mechanics and variational integrators
- Gaussian kernels on nonsimply connected closed Riemannian manifolds are never positive definite
- General techniques for constructing variational integrators
- Geometric mechanics. Part I: Dynamics and symmetry.
- Hamiltonian Hopf bifurcation with symmetry
- Hanson-Wright inequality and sub-Gaussian concentration
- scientific article; zbMATH DE number 5596589 (Why is no real title available?)
- scientific article; zbMATH DE number 3610787 (Why is no real title available?)
- scientific article; zbMATH DE number 1300852 (Why is no real title available?)
- Intrinsic statistics on Riemannian manifolds: Basic tools for geometric measurements
- Invariant variational problems
- Learning from examples as an inverse problem
- Learning particle swarming models from data with Gaussian processes
- Lie-Poisson neural networks (LPNets): data-based computing of Hamiltonian systems with symmetries
- Locally-symplectic neural networks for learning volume-preserving dynamics
- Momentum maps and Hamiltonian reduction
- Nonparametric inference of interaction laws in systems of agents from trajectory data
- On the convergence of physics informed neural networks for linear second-order elliptic and parabolic type PDEs
- On the influence of the kernel on the consistency of support vector machines
- Persistance d’orbites périodiques relatives dans les systèmes hamiltoniens symétriques
- Persistence and smoothness of critical relative elements in Hamiltonian systems with symmetry
- Persistence and stability of relative equilibria
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Quantification géométrique.
- Reduction of symplectic manifolds with symmetry
- Reproducing kernels of Sobolev spaces on \(\mathbb{R}^d\) and applications to embedding constants and tractability
- Reproducing properties of differentiable Mercer-like kernels
- Separability of reproducing kernel spaces
- Some properties of Gaussian reproducing kernel Hilbert spaces and their implications for function approximation and learning theory
- Stability and drift of underwater vehicle dynamics: mechanical systems with rigid motion symmetry
- Stochastic analysis on manifolds
- Sums and Gaussian vectors
- Symplectic Gaussian process regression of maps in Hamiltonian systems
- Symplectic learning for Hamiltonian neural networks
- Sympnets: intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems
- The Existence of Complete Riemannian Metrics
- The Gaussian kernel on the circle and spaces that admit isometric embeddings of the circle
- The local structure of Poisson manifolds
- Theorie der Transformationsgruppen. Dritter (und letzten) Abschnitt. Unter Mitwirkung von Fr. Engel bearbeitet.
- Theory of Reproducing Kernels
- Universal kernels
This page was built for publication: A global structure-preserving kernel method for the learning of Poisson systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6961695)
