Metraipolis › archive › dc_d2c92a86282a
ProtocolMatch: Protocol-Dependent Model Selection for Scientific Dynamics Forecasting — arXiv:2610.10239
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.10239
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T12:36:11.925Z
- SHA-256 of the archived text
4d366ddba18694cd3011d488cd72a3884808814ae446bc94aa7b163753020b9c- Stable URL of this page
https://metraipolis.com/documento/dc_d2c92a86282a
How to cite
ProtocolMatch: Protocol-Dependent Model Selection for Scientific Dynamics Forecasting — arXiv:2610.10239 — arXiv (en), https://arxiv.org/abs/2610.10239 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2c92a86282a (Metraipolis archive dc_d2c92a86282a, sha256 4d366ddba18694cd3011d488cd72a3884808814ae446bc94aa7b163753020b9c)
BibTeX
@misc{dcd2c92a86282a,
title = {ProtocolMatch: Protocol-Dependent Model Selection for Scientific Dynamics Forecasting — arXiv:2610.10239},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.10239},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2c92a86282a; sha256 4d366ddba18694cd3011d488cd72a3884808814ae446bc94aa7b163753020b9c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10239. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Scientific dynamics forecasting is often framed as an architecture choice, although deployment is also determined by observed history, rollout feedback, compute budget, physical objective, and test distribution. We formulate protocol-dependent model selection and introduce ProtocolMatch, a compute-matched, validation-selected, and failure-preserving evaluation framework. On driven quantum-spin dynamics, we compare recurrent, patched-attention, causal-attention, and low-rank linear predictors across three independently generated datasets. The causal-attention--recurrence ordering reverses as the training set grows within a fixed two-spin task, while a linear predictor has the lowest mean error in the six-spin local-observable comparison. Restricting observed history worsens every refreshed-history view but improves every closed-loop view in the four-spin study. A latest-state MLP has lower error than persistence on every dataset under state refresh across all five cells, yet its closed-loop rank varies by system and includes finite explosive errors. Physical penalties improve targeted consistency without reliably improving prediction error, and in-distribution intervals lose most coverage after a driving-frequency shift. Thus scientific model selection should return a predictor with its protocol and report accuracy, physical validity, and shifted-distribution reliability separately. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Quantum Physics (quant-ph). Pagina del lavoro: https://arxiv.org/abs/2610.10239 — 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 2026-10-08T12:14:15.318Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).