A closed-form structure-preserving kernel ridge regression estimator learns the Hamiltonian of a Poisson system up to a Casimir function, with PAC bounds, demonstrated on rigid body, underwater vehicle, and two-vortex dynamics.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A global structure-preserving kernel method for the learning of Poisson systems
A closed-form structure-preserving kernel ridge regression estimator learns the Hamiltonian of a Poisson system up to a Casimir function, with PAC bounds, demonstrated on rigid body, underwater vehicle, and two-vortex dynamics.