Metraipolis › archive › dc_74ecf1e7683e
Beyond Simulation: Retain-and-Repair Neural Operators for Real-World Adaptation — arXiv:2609.39387
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.39387
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:46:08.642Z
- SHA-256 of the archived text
02f98467ddcfa0107ab2e6e144e4c84047221722745f1930a2119592a404859e- Stable URL of this page
https://metraipolis.com/documento/dc_74ecf1e7683e
How to cite
Beyond Simulation: Retain-and-Repair Neural Operators for Real-World Adaptation — arXiv:2609.39387 — arXiv (en), https://arxiv.org/abs/2609.39387 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_74ecf1e7683e (Metraipolis archive dc_74ecf1e7683e, sha256 02f98467ddcfa0107ab2e6e144e4c84047221722745f1930a2119592a404859e)
BibTeX
@misc{dc74ecf1e7683e,
title = {Beyond Simulation: Retain-and-Repair Neural Operators for Real-World Adaptation — arXiv:2609.39387},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.39387},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_74ecf1e7683e; sha256 02f98467ddcfa0107ab2e6e144e4c84047221722745f1930a2119592a404859e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39387. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Neural operators increasingly benefit from pretraining on numerical simulations, yet adapting them for real-world prediction remains challenging. We introduce the Retain-and-Repair Neural Operator (R$^2$NO), a framework for adapting simulation-pretrained operators to real-world data while retaining useful pretrained structure. The pretrained operator is first finetuned on real data and then frozen to provide a source prediction, and a shared repair module learns a sequence of refinements from the same observations. Using orthogonal Fourier projections, a spectral ensemble fits a small ridge regression within each cell of the Fourier domain and combines the refinements by weights fitted on a held-out split of the real data. The cells are defined jointly by radial ranges, angular sectors, and measured channels, allowing refinement depth to vary with frequency magnitude, with orientation, and across channels. Including the source prediction as a candidate makes retention available in every cell, and independently trained repair modules enter the same combination as additional candidates. On all RealPDEBench systems and six backbones, R$^2$NO consistently outperforms full finetuning and iterative refinement. The framework treats adaptation depth as a cell-specific choice learned from real data. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.39387 — 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 2026-10-07T13:18:10.374Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).