Preprint · 2026 · Open access
Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver
arXiv (Cornell University) · Published 21 September 2026 · DOI 10.48550/arxiv.2609.24021
In brief
“Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver” is a 2026 preprint by Daniel Zhengyu Huang and Andrew M. Stuart, published in arXiv (Cornell University). OpenAlex counts 3 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).
- Citations
- 3
- Most cited in
- 2026 (3)
- References
- 98
- Authors
- 2
Abstract
Neural operators are data-driven models that learn mappings from inputs that parameterize partial differential equations, such as spatially varying coefficients, initial conditions, forcing terms, boundary conditions, or geometries, to solution fields or quantities of interest. Once trained, they can serve as surrogates for classical numerical solvers in many-query settings that require repeated evaluations for varying inputs. We address the question of when, and then why, neural operator surrogates outperform classical numerical solvers, in terms of cost for a given accuracy. We focus on the post-training, many-query limit, in which data-acquisition and training costs are treated as fixed and fully amortized. Even in this deliberately favorable regime for neural operators, there are regimes in which classical solvers outperform the surrogate models. We compare the cost-accuracy performance of neural operator surrogates and classical numerical solvers through a reproducible benchmark study comparing neural operators with problem-matched classical solvers on representative problems in computational science and engineering, focusing on prediction error, per-query floating-point cost, and wall-clock runtime. Neural operators are most competitive at low-to-moderate accuracy requirements. Their floating-point cost advantage depends strongly on the problem structure, arising when they avoid temporal or nonlinear iterations or predict a reduced quantity of interest rather than a full solution field. Additional wall-clock speedups result from dense tensor operations that are well suited to modern hardware. As the target accuracy is tightened, achieving the required accuracy with neural operators becomes increasingly challenging, and classical solvers outperform surrogates in this regime; thus classical solvers will remain important for verification and high-accuracy computation.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Huang, D. Z., & Stuart, A. M. (2026). Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2609.24021
MLA 9
Huang, Daniel Zhengyu, and Andrew M. Stuart. “Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2609.24021.
Chicago
Huang, Daniel Zhengyu and Andrew M. Stuart. 2026. “Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2609.24021.
Harvard
Huang, D. Z. and Stuart, A. M. (2026) ‘Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver’, arXiv (Cornell University). doi:10.48550/arxiv.2609.24021.
IEEE
D. Z. Huang and A. M. Stuart, “Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2609.24021.
BibTeX
@misc{huang2026cost,
title = {{Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver}},
author = {Huang, Daniel Zhengyu and Stuart, Andrew M.},
howpublished = {arXiv (Cornell University)},
year = {2026},
doi = {10.48550/arxiv.2609.24021},
url = {https://doi.org/10.48550/arxiv.2609.24021}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Model Reduction and Neural Networks
- neural operators
- surrogate models
- accuracy-cost trade-off
- partial differential equations
- prediction error
- tensor operations
Work with this paper
Most-cited papers in Physics and Astronomy
- Generalized Gradient Approximation Made SimpleJohn P. Perdew, Kieron Burke, Matthias Ernzerhof · Physical Review Letters · 1996 · 217,345 citations
- Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis setGeorg Kresse, Jürgen Furthmüller · Physical review. B, Condensed matter · 1996 · 125,164 citations
- Density-functional thermochemistry. III. The role of exact exchangeAxel D. Becke · The Journal of Chemical Physics · 1993 · 104,772 citations
- Development of the Colle-Salvetti correlation-energy formula into a functional of the electron densityChengteh Lee, Weitao Yang, Robert G. Parr · Physical review. B, Condensed matter · 1988 · 101,730 citations
- Projector augmented-wave methodPeter E. Blöchl · Physical review. B, Condensed matter · 1994 · 93,520 citations
Questions about this paper
How do I cite this paper in APA style?
Huang, D. Z., & Stuart, A. M. (2026). Cost-Accuracy Trade-offs: Neural Operator vs Classical Numerical Solver. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2609.24021
How many times has this paper been cited?
3 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Daniel Zhengyu Huang, Andrew M. Stuart.
What is the DOI of this paper?
10.48550/arxiv.2609.24021, permanently at https://doi.org/10.48550/arxiv.2609.24021. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.