Metraipolis › archive › dc_489ce1ef924f

Riemannian Neural Hamiltonian Flows: Geodesic Symplectic Transport and Interpretability — arXiv:2609.21647

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21647
License (as archived)
CC0-1.0
Archived
2026-10-10T01:45:10.043Z
SHA-256 of the archived text
d77a09359c182cc79836dc902c8b8105137bd5acbef9009210dbda591b2816b2
Stable URL of this page
https://metraipolis.com/documento/dc_489ce1ef924f

How to cite

Riemannian Neural Hamiltonian Flows: Geodesic Symplectic Transport and Interpretability — arXiv:2609.21647 — arXiv (en), https://arxiv.org/abs/2609.21647 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_489ce1ef924f (Metraipolis archive dc_489ce1ef924f, sha256 d77a09359c182cc79836dc902c8b8105137bd5acbef9009210dbda591b2816b2)
BibTeX
@misc{dc489ce1ef924f,
  title = {Riemannian Neural Hamiltonian Flows: Geodesic Symplectic Transport and Interpretability — arXiv:2609.21647},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21647},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_489ce1ef924f; sha256 d77a09359c182cc79836dc902c8b8105137bd5acbef9009210dbda591b2816b2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21647. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hamiltonian normalizing flows are attractive generative models because their phase-space maps are invertible and volume preserving, but most neural constructions are formulated in Euclidean space. We introduce Riemannian Neural Hamiltonian Flows, which combine the fixed kinetic energy of a Riemannian manifold, a learned scalar potential, and an explicit geodesic leapfrog integrator. Our analysis explains how the learned Hamiltonian can be made interpretable. Every normalizable potential defines an implicit profile, and the position marginal initially accelerates along the relative score between that profile and the base. The matched potential is the interpretable specialization for which the implicit profile is the target. In the isotropic Gaussian case, the mechanism corresponds to a phase-space rotation. A local harmonic analysis extends this result around each mode of a general target on a manifold. The gap between the learned and the matched potential is the sum of a residual memory of the base and a bias of the model, and the two potentials agree when the position base has been transferred to the momentum. This can be achieved when the former is broader than the target. Numerical experiments on Euclidean, hyperbolic, and spherical spaces show competitive sample quality and numerical cost against a Riemannian continuous normalizing flow, and confirm the interpretability of the learned potential.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-21 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.21647 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T01:14:16.640Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).