Metraipolis › archive › dc_c90a4b0ea402

Edge Accuracy Is Not Enough: Why Dynamics-Learned Structure Fails to Transfer to Inverse Problems — arXiv:2610.10213

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10213
License (as archived)
CC0-1.0
Archived
2026-10-08T12:36:09.655Z
SHA-256 of the archived text
33bdc082a4ed7b247e070c7b04e93cf04e347b935b22e98b9d783428d7e4e4b4
Stable URL of this page
https://metraipolis.com/documento/dc_c90a4b0ea402

How to cite

Edge Accuracy Is Not Enough: Why Dynamics-Learned Structure Fails to Transfer to Inverse Problems — arXiv:2610.10213 — arXiv (en), https://arxiv.org/abs/2610.10213 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c90a4b0ea402 (Metraipolis archive dc_c90a4b0ea402, sha256 33bdc082a4ed7b247e070c7b04e93cf04e347b935b22e98b9d783428d7e4e4b4)
BibTeX
@misc{dcc90a4b0ea402,
  title = {Edge Accuracy Is Not Enough: Why Dynamics-Learned Structure Fails to Transfer to Inverse Problems — arXiv:2610.10213},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10213},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c90a4b0ea402; sha256 33bdc082a4ed7b247e070c7b04e93cf04e347b935b22e98b9d783428d7e4e4b4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10213. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A natural strategy for inverse problems with scarce labelled data is to transfer relational structure learned from abundant forward-simulation data. We show this strategy fails systematically, even when it satisfies the standard theoretical justification for why structure should help. We prove that approximate structure provides estimation-error benefits whenever the edge error satisfies $\Delta < n^2 - kn$, reducing sample complexity from $O(n^2)$ to $O(kn+\Delta)$. Structure learned via Neural Relational Inference (NRI) from dynamics prediction satisfies this condition, yet on a source-localisation task across 180 CFD-simulated hydrogen-leak scenarios and 180 acoustic scenarios, it degrades performance by 116% and 201% relative to a flexible, task-optimised attention baseline, while a physics-based prior (Green's function) degrades by only 69-72%. Four independent lines of evidence show this is not a tuning failure: NRI improves only 0.5% when given 18x more training data (versus 16.6% for the task-optimised baseline, $p<0.001$); performance is insensitive to the NRI edge threshold across a wide range; the dynamics-learned graph overlaps the task-optimal graph on only 6% of edges; and two further dynamics-derived structure estimators (correlation- and mutual-information-based) show no measurable benefit over a structure-free baseline, with the correlation-based estimator performing markedly worse. We formalise this gap as a statement about approximation error that the edge-accuracy condition cannot control, and we provide a lightweight transferability test (Jaccard similarity against a partially-observed target-task graph) that separates successful from failed transfer in all four domain/structure pairs we evaluate, using under an hour of computation and 15-20% of target-domain data; we present this as a heuristic calibrated on few cases, not a validated general threshold.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.10213 — 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 2026-10-08T12:14:15.314Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).