Metraipolis › archive › dc_f24b368fc0d0
NestyNet. I. Physics Functions Are Hard to Fit with Neural Networks: A Framework for Accurate Surrogates and Analytic Derivatives — arXiv:2608.05862
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05862
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T11:02:03.637Z
- SHA-256 of the archived text
5db697f600bcc3ad611d3884b708e4ab0fc0e60e1b8a7b69d90a858a5227e73c- Stable URL of this page
https://metraipolis.com/documento/dc_f24b368fc0d0
How to cite
NestyNet. I. Physics Functions Are Hard to Fit with Neural Networks: A Framework for Accurate Surrogates and Analytic Derivatives — arXiv:2608.05862 — arXiv (en), https://arxiv.org/abs/2608.05862 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f24b368fc0d0 (Metraipolis archive dc_f24b368fc0d0, sha256 5db697f600bcc3ad611d3884b708e4ab0fc0e60e1b8a7b69d90a858a5227e73c)
BibTeX
@misc{dcf24b368fc0d0,
title = {NestyNet. I. Physics Functions Are Hard to Fit with Neural Networks: A Framework for Accurate Surrogates and Analytic Derivatives — arXiv:2608.05862},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05862},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f24b368fc0d0; sha256 5db697f600bcc3ad611d3884b708e4ab0fc0e60e1b8a7b69d90a858a5227e73c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05862. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Many of the smooth functions that matter most in physics are precisely the ones that standard neural network methods struggle to fit accurately. Here we present NestyNet, a coupled model-and-optimizer framework capable of fitting such targets to high accuracy while also delivering their gradients, Hessians, Laplacians, and antiderivatives analytically and at low cost. This makes it a natural substrate for scientific machine learning tasks. The model is a deterministic segmented analytic surrogate, and its optimizer is a second-order Levenberg--Marquardt scheme whose damping and linear solves are tailored to the stiff, strongly correlated parameter geometries induced by multiscale and sharply structured targets typical in physics and other scientific applications. On the AI Feynman benchmark of 120 physics equations, NestyNet achieves median improvement factors of $2\,100\times$ for function values, $1\,400\times$ for first derivatives, and $780\times$ for second derivatives relative to standard neural networks trained with first-order optimization (Adam). Even after refining those fits with quasi-Newton (L-BFGS) optimization, the corresponding improvements are $540\times$, $450\times$, and $250\times$. Owing to the analytic design it is up to $\approx 44\times$ faster than vectorized automatic-differentiation (autograd) baselines, with a margin growing with model size. The same analytic-derivative framework also supports vector- and complex-valued targets, measurement uncertainties in both inputs and outputs, and constraints, and its modules can be composed flexibly to build scientifically useful model architectures, all without reverting to autograd. Together, these components provide a practical modular framework for fitting difficult scientific surrogates while delivering accurate differential operators for subsequent analysis. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in physics.data-an). Categorie arXiv: Instrumentation and Methods for Astrophysics (astro-ph.IM) ; Data Analysis, Statistics and Probability (physics.data-an). Pagina del lavoro: https://arxiv.org/abs/2608.05862 — 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 2026-10-31T10:51:16.685Z dalla pagina https://arxiv.org/catchup/physics.data-an/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).