Metraipolis › archive › dc_d79976522363

Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations — arXiv:2609.34627

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34627
License (as archived)
CC0-1.0
Archived
2026-10-07T22:38:29.712Z
SHA-256 of the archived text
001685c98593860b33ed8b6e2eb09793c4b3792f0ad54e625157fc0994368c1a
Stable URL of this page
https://metraipolis.com/documento/dc_d79976522363

How to cite

Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations — arXiv:2609.34627 — arXiv (en), https://arxiv.org/abs/2609.34627 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d79976522363 (Metraipolis archive dc_d79976522363, sha256 001685c98593860b33ed8b6e2eb09793c4b3792f0ad54e625157fc0994368c1a)
BibTeX
@misc{dcd79976522363,
  title = {Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations — arXiv:2609.34627},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34627},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d79976522363; sha256 001685c98593860b33ed8b6e2eb09793c4b3792f0ad54e625157fc0994368c1a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34627. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine-learning models for physical systems are currently evaluated primarily through errors between predicted and reference states and, increasingly, through tests of physical consistency. These metrics assess whether predictions are accurate and satisfy selected physical requirements, but provide limited insight into whether learned trajectories reproduce the underlying dynamics. Domain experts examine such relationships through physical representations that expose relevant processes, interactions, and responses, but these analyses are often separated from typical machine-learning evaluation. We introduce a practical framework for evaluating dynamical fidelity through predictive structure in physical representation spaces. Experts define the representations, while reference trajectories determine which relationships are predictive and retained as evaluation tests. We demonstrate the approach in atmospheric forecasting using ERA5 representations of planetary-wave activity and Northern Annular Mode evolution, and evaluate Pangu-Weather, GraphCast, and FengWu. The models exhibit distinct departures from reference predictive structure that are not reflected by conventional forecast errors. The framework thereby turns domain-expert representations into systematic tests of learned physical dynamics without prescribing the relationships in advance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.34627 — 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 2026-10-07T22:14:28.252Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).