Metraipolis › archive › dc_bdb718c0c9df

Feature tracking in physics-informed neural networks via joint optimization of nonlinear deformation manifolds: application to shocks — arXiv:2610.02230

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.02230
License (as archived)
CC0-1.0
Archived
2026-10-06T10:02:28.591Z
SHA-256 of the archived text
6d0d15317c338ca25b727be41b991d8f017798921c9205a9fb2977b5d859ba16
Stable URL of this page
https://metraipolis.com/documento/dc_bdb718c0c9df

How to cite

Feature tracking in physics-informed neural networks via joint optimization of nonlinear deformation manifolds: application to shocks — arXiv:2610.02230 — arXiv (en), https://arxiv.org/abs/2610.02230 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bdb718c0c9df (Metraipolis archive dc_bdb718c0c9df, sha256 6d0d15317c338ca25b727be41b991d8f017798921c9205a9fb2977b5d859ba16)
BibTeX
@misc{dcbdb718c0c9df,
  title = {Feature tracking in physics-informed neural networks via joint optimization of nonlinear deformation manifolds: application to shocks — arXiv:2610.02230},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.02230},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bdb718c0c9df; sha256 6d0d15317c338ca25b727be41b991d8f017798921c9205a9fb2977b5d859ba16},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.02230. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Physics-informed neural networks (PINNs) often converge to inaccurate solutions for conservation laws with shocks, because uniformly distributed collocation points undersample localized features and let the residual be dominated by regions that are already well resolved. We propose a feature-tracking PINN (FT-PINN) in which the solution network is defined on a fixed reference domain and composed with a diffeomorphic deformation map from a parameterized nonlinear manifold. The deformation and solution-network parameters are trained jointly by minimizing the pulled-back conservation-law residual. This lets collocation points concentrate along features of essentially arbitrary geometry, including curved, oblique, and merging shocks, without prior knowledge of their locations. The framework is agnostic to the choice of parameterization. Boundary preservation is enforced exactly through a tangential projection of the displacement, and folding is discouraged by a one-sided penalty on the Jacobian determinant. On four test problems (space-time viscous Burgers with merging shocks, a decelerating Burgers shock, the space-time Euler shock tube, and steady 2D Euler regular shock reflection), FT-PINN resolves shocks at their correct locations with a limited collocation budget. A vanilla PINN with the same architecture, budget, and training either misplaces the shocks or fails to form them.

Tema degli esploratori: fisica e astronomia (retro-catalogo arXiv: annunci del 2026-10-05 in physics.comp-ph).
Categorie arXiv: Computational Engineering, Finance, and Science (cs.CE) ; Computational Physics (physics.comp-ph); Fluid Dynamics (physics.flu-dyn); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.02230 — 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 2026-10-06T09:44:18.322Z dalla pagina https://arxiv.org/catchup/physics.comp-ph/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).