Abstract
In recent works, the authors introduced a neural operator: a special type of neural networks that can approximate maps between infinite-dimensional spaces. Using numerical and analytical techniques, we will highlight the peculiarities of the training and evaluation of these operators. In particular, we will show that, for a broad class of neural operators based on integral transforms, a systematic bias is inevitable, owning to aliasing errors. To avoid this bias, we introduce spectral neural operators based on explicit discretization of the domain and the codomain. Although discretization deteriorates the approximation properties, numerical experiments show that the accuracy of spectral neural operators is often superior to the one of neural operators defined on infinite-dimensional Banach spaces.
Similar content being viewed by others
REFERENCES
T. E. Hull et al., “Comparing numerical methods for ordinary differential equations,” SIAM J. Numer. Anal. 9, 603–637 (1972).
B. Fornberg and G. B. Whitham, “A numerical and theoretical study of certain nonlinear wave phenomena,” Philos. Trans. R. Soc. London Ser. A: Math. Phys. Sci. 289 (1361), 373–404 (1978).
T. Yoneyama, “The Korteweg–de Vries two-soliton solution as interacting two single solitons,” Prog. Theor. Phys. 71 (4), 843–846 (1984).
I. E. Lagaris, A. Likas, and D. I. Fotiadis, “Artificial neural networks for solving ordinary and partial differential equations,” IEEE Trans. Neural Networks 9 (5), 987–1000 (1998).
L. N. Trefethen, Spectral Methods in MATLAB (SIAM, Philadelphia, 2000).
J. P. Boyd, Chebyshev and Fourier Spectral Methods (Dover, Mineola, NY, 2001).
M. L. Saksman, S. Siltanen, et al., “Discretization-invariant Bayesian inversion and Besov space priors” (2009). https://doi.org/10.48550/arXiv.0901.4220
D. J. Kedziora, A. Ankiewicz, and N. Akhmediev, “Second-order nonlinear Schrödinger equation breather solutions in the degenerate and rogue wave limits,” Phys. Rev. E 85 (6), 066601 (2012).
Y. Bar-Sinai et al., “Learning data-driven discretizations for partial differential equations,” Proc. Natl. Acad. Sci. 116, 15344–15349 (2019).
L. Lu, P. Jin, and G. E. Karniadakis, “DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators” (2019). https://doi.org/10.48550/arXiv.1910.03193
L. N. Trefethen, Approximation Theory and Approximation Practice (SIAM, 2019).
B. Ummenhofer et al., “Lagrangian fluid simulation with continuous convolutions,” in International Conference on Learning Representations (2019).
Z. Li et al., “Fourier neural operator for parametric partial differential equations” (2020). https://doi.org/10.48550/arXiv.2010.08895
Z. Li et al., “Neural operator: Graph kernel network for partial differential equations” (2020). https://doi.org/10.48550/arXiv.2003.03485
G. Gupta, X. Xiao, and P. Bogdan, “Multiwavelet-based operator learning for differential equations,” Adv. Neural Inf. Process. Syst. 34, 24048–24062 (2021). https://doi.org/10.48550/arXiv.2109.13459
N. Kovachki et al., “Neural operator: Learning maps between function spaces” (2021). https://doi.org/10.48550/arXiv.2108.08481
T. Tripura and S. Chakraborty, “Wavelet neural operator: A neural operator for parametric partial differential equations” (2022). https://arxiv.org/abs/2205.02191
Funding
This work was supported by the Ministry of Science and Higher Education of the Russian Federation (project no. 075-10-2021-068).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
The authors of this work declare that they have no conflicts of interest.
Additional information
Publisher’s Note.
Pleiades Publishing remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Fanaskov, V.S., Oseledets, I.V. Spectral Neural Operators. Dokl. Math. 108 (Suppl 2), S226–S232 (2023). https://doi.org/10.1134/S1064562423701107
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1064562423701107