Metraipolis › archive › dc_9b1da00358ca

Unrolled gradients in disguise: bridging interpolation-based and Jacobian regularization for stable neural dynamics — arXiv:2611.03461

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03461
License (as archived)
CC0-1.0
Archived
2026-11-03T18:39:04.562Z
SHA-256 of the archived text
add43d57f7afcbd8a17f7552f21911edb92f3d517152ae5f0d79fcc2906a7e32
Stable URL of this page
https://metraipolis.com/documento/dc_9b1da00358ca

How to cite

Unrolled gradients in disguise: bridging interpolation-based and Jacobian regularization for stable neural dynamics — arXiv:2611.03461 — arXiv (en), https://arxiv.org/abs/2611.03461 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9b1da00358ca (Metraipolis archive dc_9b1da00358ca, sha256 add43d57f7afcbd8a17f7552f21911edb92f3d517152ae5f0d79fcc2906a7e32)
BibTeX
@misc{dc9b1da00358ca,
  title = {Unrolled gradients in disguise: bridging interpolation-based and Jacobian regularization for stable neural dynamics — arXiv:2611.03461},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03461},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9b1da00358ca; sha256 add43d57f7afcbd8a17f7552f21911edb92f3d517152ae5f0d79fcc2906a7e32},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03461. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hybrid models and neural differential equations (NDEs) are becoming increasingly important for modelling physical systems; however, they often encounter stability and accuracy issues during long-term integration. Training on unrolled trajectories is known to limit these divergences but quickly becomes too computationally expensive due to the need to compute gradients over an iterative process, making this method impractical for large-scale physical systems (e.g. weather and climate forecasting). Two popular paradigms for improving the stability of learned dynamics are to train on interpolated or noisy samples or to regularize the Jacobian of the learned representation. In this work, we demonstrate that these paradigms can be understood under a common lens and that they both correspond to an approximation of the gradients obtained in an unrolled learning strategy. Building on this understanding, we derive a principled and complete approach to regularize training and demonstrate its impact on the stability of learned dynamical systems.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Numerical Analysis (math.NA).
Pagina del lavoro: https://arxiv.org/abs/2611.03461 — 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 2026-11-03T18:12:54.644Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).