Metraipolis › archive › dc_ff011344414e
Staying on the Attractor: Supervising Neural Surrogates of 3D Turbulence Where They Leave It — arXiv:2609.32864
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32864
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:36:55.682Z
- SHA-256 of the archived text
a62c15273361efcd363c6e3bc4869e71648ec87a369302dc8e0762cd37fbb450- Stable URL of this page
https://metraipolis.com/documento/dc_ff011344414e
How to cite
Staying on the Attractor: Supervising Neural Surrogates of 3D Turbulence Where They Leave It — arXiv:2609.32864 — arXiv (en), https://arxiv.org/abs/2609.32864 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff011344414e (Metraipolis archive dc_ff011344414e, sha256 a62c15273361efcd363c6e3bc4869e71648ec87a369302dc8e0762cd37fbb450)
BibTeX
@misc{dcff011344414e,
title = {Staying on the Attractor: Supervising Neural Surrogates of 3D Turbulence Where They Leave It — arXiv:2609.32864},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32864},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff011344414e; sha256 a62c15273361efcd363c6e3bc4869e71648ec87a369302dc8e0762cd37fbb450},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32864. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Neural surrogates are trained to predict 3D turbulent flows in place of direct numerical simulation (DNS). For chaotic flows, the goal is short-term pointwise accuracy followed by long-term physical and statistical fidelity. However, small prediction errors can carry a surrogate away from the flow's attractor. Off-attractor states are poorly represented in training data, leaving their evolution weakly constrained. The learned dynamics can then amplify deviations and lead to blow-up, freezing, or statistical drift. We propose off-attractor supervision (OAS) to supervise neural surrogates where they leave the attractor. OAS teaches the model how the true Navier-Stokes dynamics would evolve from these states. Each selected state is paired with its own future computed by DNS. Three generators select a few hundred states for relabeling. The first collects states from the surrogate's own rollouts. The second uses surrogate attacks to target freezing, excessive amplification, and violations of incompressibility and energy balance. The third perturbs training states along an amplified direction and a strongly damped random direction of the dynamics. All attacks run on the surrogate alone, and DNS relabeling is performed offline once per selected state. Experiments on $128^3$ turbulence show that OAS increases the median time to failure from 21 to 721 steps. The compared baselines achieve medians of at most 110 steps, and the advantage holds across training seeds. OAS also achieves the lowest pointwise error at step 15 and the best long-horizon statistics among the compared methods. OAS integrates physical models into neural simulation by extending supervision from fixed reference trajectories to states where the surrogate is likely to fail. This principle can guide the development of more reliable scientific surrogates when deployment takes models beyond the coverage of their training data. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG) ; Fluid Dynamics (physics.flu-dyn). Pagina del lavoro: https://arxiv.org/abs/2609.32864 — 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 2026-10-07T22:14:27.791Z 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).