Metraipolis › archive › dc_d56bae145c52
Latent Twin Operator — arXiv:2609.35531
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35531
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:25:24.448Z
- SHA-256 of the archived text
fb72d4e33730e6c1e214350db1ecaeb6368bc5ef557cc175816221d1a684a814- Stable URL of this page
https://metraipolis.com/documento/dc_d56bae145c52
How to cite
Latent Twin Operator — arXiv:2609.35531 — arXiv (en), https://arxiv.org/abs/2609.35531 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d56bae145c52 (Metraipolis archive dc_d56bae145c52, sha256 fb72d4e33730e6c1e214350db1ecaeb6368bc5ef557cc175816221d1a684a814)
BibTeX
@misc{dcd56bae145c52,
title = {Latent Twin Operator — arXiv:2609.35531},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35531},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d56bae145c52; sha256 fb72d4e33730e6c1e214350db1ecaeb6368bc5ef557cc175816221d1a684a814},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35531. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Surrogate models deployed on real physical systems rarely see data at fixed resolutions: sensor configurations vary across deployments and may evolve over time as sensing infrastructure changes. We introduce the Latent Twin Operator (LTO), a latent-space surrogate for time-evolving PDEs with a Convolutional Conditional Neural Process (ConvCNP)-style encoder and decoder that accepts a context set of $N$ sensor observations with arbitrary placement and can be queried at any resolution. A learned latent evolution map advances the encoded state directly between arbitrary time points, decoupling temporal evolution from the observation and query discretizations. We derive an explicit $\mathcal{O}(N^{-2/(3D)})$ rate for the context discretization error in spatial dimension $D$ under quasi-uniform refinement. We verify the predicted decay empirically on a 2D heat-equation benchmark. Across time-dependent PDE benchmarks, LTO achieves strong accuracy under one-step comparisons and transfers from native to coarser spatial resolutions with fixed parameters. On Navier--Stokes, its direct latent evolution further reduces error over longer prediction horizons relative to recursive evaluation.
Tema degli esploratori: matematica e informatica teorica (retro-catalogo arXiv: annunci del 2026-09-29 in math.NA).
Categorie arXiv: Computational Engineering, Finance, and Science (cs.CE) ; Numerical Analysis (math.NA).
Pagina del lavoro: https://arxiv.org/abs/2609.35531 — 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 2026-10-07T23:04:09.542Z dalla pagina https://arxiv.org/catchup/math.NA/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).