Metraipolis › archive › dc_bf80bada88c5
Learning When to Refine: Long-Horizon Reinforcement Learning for Budgeted Neural-Operator PDE Solvers — arXiv:2610.06883
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.06883
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:19:59.012Z
- SHA-256 of the archived text
d09e32466956b9354effeca2afa14afe9f64c33f0ba1e46f090821e1dcc27e4b- Stable URL of this page
https://metraipolis.com/documento/dc_bf80bada88c5
How to cite
Learning When to Refine: Long-Horizon Reinforcement Learning for Budgeted Neural-Operator PDE Solvers — arXiv:2610.06883 — arXiv (en), https://arxiv.org/abs/2610.06883 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf80bada88c5 (Metraipolis archive dc_bf80bada88c5, sha256 d09e32466956b9354effeca2afa14afe9f64c33f0ba1e46f090821e1dcc27e4b)
BibTeX
@misc{dcbf80bada88c5,
title = {Learning When to Refine: Long-Horizon Reinforcement Learning for Budgeted Neural-Operator PDE Solvers — arXiv:2610.06883},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.06883},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf80bada88c5; sha256 d09e32466956b9354effeca2afa14afe9f64c33f0ba1e46f090821e1dcc27e4b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.06883. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Neural operators provide fast surrogates for time-dependent PDEs, but autoregressive deployment creates a refinement-allocation problem: prediction errors vary over space and time, while only a finite number of local corrections can be committed along a trajectory. We formulate this as budgeted adaptive neural-operator solving. A global Fourier neural operator advances the full field, a local operator proposes patch-wise residual corrections, and a set-aware selector chooses where to refine. A macro policy decides when and how much of the remaining refinement budget to spend. We introduce rollout-verified policy improvement (RV-PI), which evaluates feasible refinement counts through actual continuation rollouts of the learned PDE solver, converts long-horizon advantages into conservative policy targets, and accepts an update only when held-out trajectory error improves. On the shallow-water benchmark with a 32-intervention budget, RV-PI achieves a three-seed mean trajectory relative L2 error of 0.6910, improving over immediate-only policy improvement by 5.37% and RandomMacro by 2.41%. On the forcing-driven Brusselator benchmark with a 76-intervention budget, RV-PI attains 0.09954, improving over immediate-only policy improvement by 2.31% and RandomMacro by 5.32%. These results show that, under a fixed refinement budget, the value of a local correction depends on its downstream effect on the autoregressive trajectory, not only on its immediate error reduction. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.06883 — 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 2026-10-07T11:15:24.641Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).