Pith. sign in

REVIEW 2 major objections 1 minor 43 references

Domain-Decomposed Randomized Neural Networks for Partial Differential Equations in Unbounded Domains

T0 review · 2 major / 1 minor · reviewed 2026-07-01 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Domain-decomposed randomized neural networks approximate PDE solutions on unbounded domains with a conditional error bound proved in broken Sobolev norms.

desk verdict The domain-decomposed randomized NN setup for unbounded PDEs is a practical split of near- and far-field subnetworks solved by least-squares, but the conditional approximation theorem rests on an unverified bounded-weights assumption for the far-field part. read the letter →

arxiv 2606.31342 v1 pith:DNQWEFKF submitted 2026-06-30 math.NA cs.LGcs.NA

Haixin Wang , Haoning Dang , Fei Wang , Shimin Guo This is my paper · ORCID
classification math.NAcs.LGcs.NA
keywords domaindecompositionrandomizedneuralnetworksunboundeddomainspartialdifferentialequationsleast-squaresmethodsPetrov-GalerkincollocationbrokenSobolevnorms
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that splitting an unbounded domain into near-field and far-field regions and assigning separate randomized subnetworks to each allows accurate solution of partial differential equations without artificial truncation boundaries. The subnetworks are linked only through interface conditions, with output-layer weights obtained from linear least-squares problems derived from either Petrov-Galerkin or collocation discretizations. A conditional bounded-parameter approximation theorem is established in a broken Sobolev norm, accompanied by an explicit decomposition of the total error into approximation, quadrature, and optimization components. Numerical tests on Poisson and time-dependent Schrödinger problems confirm that the resulting systems remain solvable and produce accurate results across different geometries and decay behaviors.

What carries the argument

Domain-decomposed randomized neural networks whose subnetworks are coupled at interfaces and whose trainable coefficients are recovered from linear least-squares systems

What would settle it

Numerical runs in which random weights are deliberately scaled to grow without bound, producing singular or severely ill-conditioned least-squares matrices and observed errors that exceed the stated bounds, would falsify the conditional result.

Watch

Extended reading notes

Core claim

Different randomized subnetworks are assigned to different spatial regimes: a near-field subnetwork captures local and geometric features, whereas a far-field subnetwork represents exterior decay; the subnetworks are coupled by boundary and interface conditions, only the output-layer coefficients are solved from linear least-squares systems, and a conditional bounded-parameter approximation result holds in a broken Sobolev norm together with an error decomposition covering approximation, empirical-consistency/quadrature, and least-squares optimization errors.

Load-bearing premise

The random weights inside each subnetwork remain bounded so the conditional approximation result applies and the coupled least-squares systems stay well-posed.

Editorial extensions

If this is right

  • A Petrov-Galerkin formulation works for semi-unbounded elliptic problems while a collocation formulation covers fully unbounded, perforated, and time-dependent problems.
  • The total error is bounded by the sum of three explicitly identified contributions: approximation error of the subnetworks, quadrature or empirical-consistency error, and optimization error from the least-squares solve.
  • The method applies directly to Poisson and time-dependent Schrödinger equations without requiring problem-specific artificial boundary conditions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same decomposition strategy could be tested on nonlinear or higher-order PDEs where near-field and far-field behaviors differ sharply.
  • If the bounded-weight premise fails in practice, adding a simple weight-regularization term to the least-squares objective would be a direct practical safeguard.
  • The approach separates local geometric resolution from global decay modeling, which may reduce the total number of trainable parameters compared with a single global network on very large domains.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 1 minor

Summary. The paper proposes a domain-decomposed randomized neural network method for PDEs on unbounded domains. Separate randomized subnetworks handle near-field (local/geometric features) and far-field (exterior decay) regimes; these are coupled only through boundary/interface conditions, with solely the output-layer coefficients determined by linear least-squares arising from Petrov-Galerkin or collocation discretizations. A conditional bounded-parameter approximation theorem is proved in a broken Sobolev norm, accompanied by an error decomposition that separates approximation, empirical-consistency/quadrature, and optimization contributions. Numerical results are shown for Poisson and time-dependent Schrödinger problems.

Significance. If the conditioning assumptions hold, the framework supplies a flexible alternative to artificial-boundary truncation or global spectral bases for problems whose near- and far-field behaviors differ markedly. The explicit error decomposition that isolates three distinct sources of error is a constructive feature that could aid future analysis.

major comments (2)
  1. [Abstract / statement of the conditional bounded-parameter approximation result] The central approximation result (abstract) is stated to be conditional on the random weights remaining bounded so that the subnetworks stay stable. No argument is supplied that the chosen sampling distributions keep the far-field subnetwork weights inside the required bound when the exterior decay must be represented on an unbounded domain; the same boundedness premise is also invoked to assert well-posedness of the interface least-squares systems. Because the error decomposition and the well-posedness claim both rest on this premise, its verification is load-bearing.
  2. [Abstract / error-decomposition paragraph] The abstract asserts that the least-squares systems arising from the coupled interface conditions are solved without post-hoc tuning, yet the manuscript supplies neither explicit constants in the error bound nor numerical verification that the random-feature matrices remain well-conditioned once the far-field subnetwork is included.
minor comments (1)
  1. The abstract refers to 'a conditional bounded-parameter approximation result' without indicating the theorem number or the precise norm in which the bound is stated.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. We address each major comment below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [Abstract / statement of the conditional bounded-parameter approximation result] The central approximation result (abstract) is stated to be conditional on the random weights remaining bounded so that the subnetworks stay stable. No argument is supplied that the chosen sampling distributions keep the far-field subnetwork weights inside the required bound when the exterior decay must be represented on an unbounded domain; the same boundedness premise is also invoked to assert well-posedness of the interface least-squares systems. Because the error decomposition and the well-posedness claim both rest on this premise, its verification is load-bearing.

    Authors: We agree that the bounded-parameter assumption is central to both the approximation theorem (Theorem 4.1) and the well-posedness of the interface least-squares problems. The manuscript selects standard sampling distributions (Gaussian or uniform) for the random weights but does not supply an explicit argument or probability bound ensuring these weights remain inside the stability threshold for the far-field subnetwork on unbounded domains. We will add a short supporting lemma in Section 3 that uses concentration inequalities to bound the probability that the far-field weights exceed the required threshold, thereby strengthening the conditional result and the well-posedness claim. revision: yes

  2. Referee: [Abstract / error-decomposition paragraph] The abstract asserts that the least-squares systems arising from the coupled interface conditions are solved without post-hoc tuning, yet the manuscript supplies neither explicit constants in the error bound nor numerical verification that the random-feature matrices remain well-conditioned once the far-field subnetwork is included.

    Authors: The abstract phrase 'solved without post-hoc tuning' indicates that, once random features are drawn, the output coefficients are obtained from a single linear least-squares solve with no additional regularization or iterative hyperparameter adjustment. We acknowledge that the error bounds in the decomposition (Section 4) are stated in terms of the Gram-matrix conditioning without fully explicit constants independent of the realization, and that numerical condition-number tables specifically for the far-field subnetwork are not reported. We will revise the manuscript to include both a brief discussion of how the conditioning enters the constants and numerical experiments reporting condition numbers of the interface matrices when the far-field subnetwork is active. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: derivation chain is self-contained

full rationale

The paper states a conditional bounded-parameter approximation result proved in a broken Sobolev norm together with an explicit error decomposition that separately accounts for approximation, empirical-consistency/quadrature, and least-squares optimization errors. No equations, fitted parameters, or self-citations are presented that would reduce any claimed bound or well-posedness statement to a tautology by construction. The domain-decomposed subnetwork assignment and coupling via interface conditions are introduced as new elements whose error analysis is carried out independently of the target quantities.

Assumptions & free parameters 0 free parameters · 0 assumptions · 0 invented entities

Only the abstract is available, so the ledger is necessarily incomplete; no explicit free parameters, axioms, or invented entities are named in the provided text.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Domain-Decomposed Randomized Neural Networks for Partial Differential Equations in Unbounded Domains." pith.science (2026). https://pith.science/paper/DNQWEFKF

@misc{pith2026260631342,
  author       = {Pith},
  title        = {Pith review of: Domain-Decomposed Randomized Neural Networks for Partial Differential Equations in Unbounded Domains},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DNQWEFKF}},
  note         = {Machine review of arXiv:2606.31342}
}
read the original abstract

Partial differential equations on unbounded domains are challenging because the exterior region must be represented without excessive truncation error. Truncation-based methods often require problem-dependent artificial boundary conditions, while global spectral bases may be inefficient for localized structures, irregular geometries, or solutions with different near-field and far-field behaviors. We propose a domain-decomposed randomized neural network framework for such problems. Different randomized subnetworks are assigned to different spatial regimes: a near-field subnetwork captures local and geometric features, whereas a far-field subnetwork represents exterior decay. The subnetworks are coupled by boundary and interface conditions, and only the output-layer coefficients are solved from linear least-squares systems arising from Petrov--Galerkin or collocation formulations. We develop a Petrov--Galerkin method for semi-unbounded elliptic problems and a collocation method for fully unbounded, perforated, and time-dependent problems. A conditional bounded-parameter approximation result is proved in a broken Sobolev norm, together with an error decomposition covering approximation, empirical-consistency/quadrature, and least-squares optimization errors. Numerical experiments for Poisson and time-dependent Schr\"odinger equations demonstrate the accuracy and flexibility of the proposed method.

Figures

Figures reproduced from arXiv: 2606.31342 by the authors.

Figure 1
Figure 1. Logarithmic errors of the DD-RaNN–PG (left) and the Laguerre-Spectral-Galerkin method (right) for [PITH_FULL_IMAGE:figures/full_fig_p030_1.png] view at source ↗
Figure 2
Figure 2. Scatter plot of sampling points (SPs) in DD-RaNN-CM (left) and logarithmic errors for DD-RaNN-CM [PITH_FULL_IMAGE:figures/full_fig_p032_2.png] view at source ↗
Figure 3
Figure 3. Computational solution obtained using the DD-RaNN-CM (left), exact solution (middle), and absolute [PITH_FULL_IMAGE:figures/full_fig_p032_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Scatter plot of sampling points (SPs) for DD-RaNN-CM (left) and the logarithmic errors for DD-RaNN [PITH_FULL_IMAGE:figures/full_fig_p034_4.png]
Figure 5
Figure 5. Figure 5: Computational solution obtained using the DD-RaNN-CM (left), exact solution (middle), and absolute [PITH_FULL_IMAGE:figures/full_fig_p034_5.png]
Figure 6
Figure 6. Figure 6: Scatter plot of sampling points (SPs) for RaNN1 and RaNN2 within a cube ranging from -10 to 10 (left) [PITH_FULL_IMAGE:figures/full_fig_p035_6.png]
Figure 7
Figure 7. Figure 7: Scatter plot of the sampling points (SPs) distribution for Example [PITH_FULL_IMAGE:figures/full_fig_p037_7.png]
Figure 8
Figure 8. Figure 8: Logarithmic errors for the real part (left) and imaginary part (right) of Example [PITH_FULL_IMAGE:figures/full_fig_p037_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 43 canonical work pages

  1. [1]

    A unified deep artificial neural network approach to partial differential equations in complex geometries.Neurocomputing, 317:28–41, 2018

    Jens Berg and Kaj Nystr¨ om. A unified deep artificial neural network approach to partial differential equations in complex geometries.Neurocomputing, 317:28–41, 2018

    work page 2018

  2. [2]

    Carolin Birk, Maximilian Reichel, and J¨ org Schr¨ oder. Magnetostatic simulations with consideration of exterior domains using the scaled boundary finite element method.Computer Methods in Applied Mechanics and Engineering, 399:115362, 2022

    work page 2022

  3. [3]

    Two-grid finite volume element method for the time-dependent Schr¨ odinger equation.Computers & Mathematics with Applications, 108:185–195, 2022

    Chuanjun Chen, Yuzhi Lou, and Hanzhang Hu. Two-grid finite volume element method for the time-dependent Schr¨ odinger equation.Computers & Mathematics with Applications, 108:185–195, 2022

    work page 2022

  4. [4]

    Bridging traditional and machine learning-based algorithms for solving PDEs: the random feature method.Journal of Machine Learning, 1(3):268–298, 2022

    Jingrun Chen, Xurong Chi, Weinan E, and Zhouwang Yang. Bridging traditional and machine learning-based algorithms for solving PDEs: the random feature method.Journal of Machine Learning, 1(3):268–298, 2022

    work page 2022

  5. [5]

    Analysis of absorbing boundary conditions for the anomalous diffusion in comb model on unbounded domain by finite volume method.Applied Mathematics Letters, 144:108712, 2023

    Siyu Chen, Lin Liu, Jiajia Li, Jingyu Yang, Libo Feng, and Jiangshan Zhang. Analysis of absorbing boundary conditions for the anomalous diffusion in comb model on unbounded domain by finite volume method.Applied Mathematics Letters, 144:108712, 2023

    work page 2023

  6. [6]

    Local randomized neural networks with hybridized discontinuous Petrov– Galerkin methods for Stokes–Darcy flows.Physics of Fluids, 36(8):087138, 2024

    Haoning Dang and Fei Wang. Local randomized neural networks with hybridized discontinuous Petrov– Galerkin methods for Stokes–Darcy flows.Physics of Fluids, 36(8):087138, 2024

    work page 2024

  7. [7]

    Adaptive-growth randomized neural networks for PDEs: Algorithms and numerical analysis.arXiv preprint arXiv:2408.17225, 2024

    Haoning Dang, Fei Wang, and Song Jiang. Adaptive-growth randomized neural networks for PDEs: Algorithms and numerical analysis.arXiv preprint arXiv:2408.17225, 2024

    work page arXiv 2024

  8. [8]

    Peng Ding, Yubin Yan, Zongqi Liang, and Yuyuan Yan. Finite difference method for time-fractional Klein– Gordon equation on an unbounded domain using artificial boundary conditions.Mathematics and Computers in Simulation, 205:902–925, 2023

    work page 2023

Show all 43 references
  1. [9]

    M. W. M. G. Dissanayake and N. Phan-Thien. Neural-network-based approximations for solving partial differential equations.Communications in Numerical Methods in Engineering, 10(3):195–201, 1994

    1994

  2. [10]

    Suchuan Dong and Zongwei Li. Local extreme learning machines and domain decomposition for solving linear and nonlinear partial differential equations.Computer Methods in Applied Mechanics and Engineering, 387:114129, 2021

    2021

  3. [11]

    The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems.Communications in Mathematics and Statistics, 6(1):1–12, 2018

    Weinan E and Bing Yu. The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems.Communications in Mathematics and Statistics, 6(1):1–12, 2018

    2018

  4. [12]

    Two FEM-BEM methods for the numerical solution of 2D transient elastodynamics problems in unbounded domains.Computers & Mathematics with Applications, 114:132–150, 2022

    Silvia Falletta, Giovanni Monegato, and Letizia Scuderi. Two FEM-BEM methods for the numerical solution of 2D transient elastodynamics problems in unbounded domains.Computers & Mathematics with Applications, 114:132–150, 2022. 39

    2022

  5. [13]

    A new absorbing layer approach for solving the nonlinear Schr¨ odinger equation

    Feng Guo and Weizhong Dai. A new absorbing layer approach for solving the nonlinear Schr¨ odinger equation. Applied Numerical Mathematics, 189:88–106, 2023

    2023

  6. [14]

    Shimin Guo, Yaping Chen, Liquan Mei, and Yining Song. Finite difference/generalized Hermite spectral method for the distributed-order time-fractional reaction-diffusion equation on multi-dimensional unbounded domains.Computers & Mathematics with Applications, 93:1–19, 2021

    2021

  7. [15]

    Dissipation-preserving rational spectral-Galerkin method for strongly damped nonlinear wave system involving mixed fractional Laplacians in unbounded domains

    Shimin Guo, Wenjing Yan, Can Li, and Liquan Mei. Dissipation-preserving rational spectral-Galerkin method for strongly damped nonlinear wave system involving mixed fractional Laplacians in unbounded domains. Journal of Scientific Computing, 93(2):53, 2022

    2022

  8. [16]

    Explicit time-domain analysis of wave propagation in unbounded domains using the scaled boundary finite element method.Engineering Analysis with Boundary Elements, 168:105891, 2024

    T Kuhn, H Gravenkamp, and C Birk. Explicit time-domain analysis of wave propagation in unbounded domains using the scaled boundary finite element method.Engineering Analysis with Boundary Elements, 168:105891, 2024

    2024

  9. [17]

    I. E. Lagaris, A. Likas, and D. I. Fotiadis. Artificial neural networks for solving ordinary and partial differential equations.IEEE Transactions on Neural Networks, 9(5):987–1000, 1998

    1998

  10. [18]

    Numerical solution of the regularized logarithmic Schr¨ odinger equation on unbounded domains.Applied Numerical Mathematics, 140:91–103, 2019

    Hongwei Li, Xin Zhao, and Yunxia Hu. Numerical solution of the regularized logarithmic Schr¨ odinger equation on unbounded domains.Applied Numerical Mathematics, 140:91–103, 2019

    2019

  11. [19]

    Local randomized neural networks with finite difference methods for interface problems.Journal of Computational Physics, 529:113847, 2025

    Yunlong Li and Fei Wang. Local randomized neural networks with finite difference methods for interface problems.Journal of Computational Physics, 529:113847, 2025

    2025

  12. [20]

    A finite element method for elliptic optimal control problem in the unbounded domain.Journal of Applied Mathematics and Computing, 71(3):4375–4396, 2025

    Jitong Lin, Yanping Chen, and Yunqing Huang. A finite element method for elliptic optimal control problem in the unbounded domain.Journal of Applied Mathematics and Computing, 71(3):4375–4396, 2025

    2025

  13. [21]

    Analysis and Hermite spectral approximation of diffusive-viscous wave equations in unbounded domains arising in geophysics.Journal of Scientific Computing, 95:51, 2023

    Dan Ling and Zhiping Mao. Analysis and Hermite spectral approximation of diffusive-viscous wave equations in unbounded domains arising in geophysics.Journal of Scientific Computing, 95:51, 2023

    2023

  14. [22]

    Integral representations of Sobolev spaces via Relu k activation function and optimal error estimates for linearized networks.arXiv preprint arXiv:2505.00351, 2025

    Xinliang Liu, Tong Mao, and Jinchao Xu. Integral representations of Sobolev spaces via Relu k activation function and optimal error estimates for linearized networks.arXiv preprint arXiv:2505.00351, 2025

    2025

  15. [23]

    Dina Mostafa, Mahmoud A Zaky, Ramy M Hafez, Ahmed S Hendy, Mohamed A Abdelkawy, and Ahmed A Aldraiweesh. Tanh Jacobi spectral collocation method for the numerical simulation of nonlinear Schr¨ odinger equations on unbounded domain.Mathematical Methods in the Applied Sciences, ...

    2023

  16. [24]

    Physics informed deep learning (part i): Data- driven solutions of nonlinear partial differential equations.arXiv preprint arXiv:1711.10561, 2017

    Maziar Raissi, Paris Perdikaris, and George Em Karniadakis. Physics informed deep learning (part i): Data- driven solutions of nonlinear partial differential equations.arXiv preprint arXiv:1711.10561, 2017

    2017 arXiv

  17. [25]

    Physics informed deep learning (part ii): Data- driven discovery of nonlinear partial differential equations.arXiv preprint arXiv:1711.10566, 2017

    Maziar Raissi, Paris Perdikaris, and George Em Karniadakis. Physics informed deep learning (part ii): Data- driven discovery of nonlinear partial differential equations.arXiv preprint arXiv:1711.10566, 2017

    2017 arXiv

  18. [26]

    Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

    Maziar Raissi, Paris Perdikaris, and George Em Karniadakis. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics, 378:686–707, 2019

    2019

  19. [27]

    Overlapping Schwarz preconditioners for randomized neural networks with domain decomposition.Computer Methods in Applied Mechanics and Engineering, 442:118011, 2025

    Yong Shang, Alexander Heinlein, Siddhartha Mishra, and Fei Wang. Overlapping Schwarz preconditioners for randomized neural networks with domain decomposition.Computer Methods in Applied Mechanics and Engineering, 442:118011, 2025

    2025

  20. [28]

    Randomized neural networks with Petrov–Galerkin methods for solving linear elasticity and Navier–Stokes equations.Journal of Engineering Mechanics, 150(4):04024010, 2024

    Yong Shang and Fei Wang. Randomized neural networks with Petrov–Galerkin methods for solving linear elasticity and Navier–Stokes equations.Journal of Engineering Mechanics, 150(4):04024010, 2024

    2024

  21. [29]

    Yong Shang, Fei Wang, and Jingbo Sun. Randomized neural network with Petrov–Galerkin methods for solving linear and nonlinear partial differential equations.Communications in Nonlinear Science and Numerical Simulation, 127:107518, 2023. 40

    2023

  22. [30]

    Some recent advances on spectral methods for unbounded domains.Communications in Computational Physics, 5(2-4):195–241, 2009

    Jie Shen and Lilian Wang. Some recent advances on spectral methods for unbounded domains.Communications in Computational Physics, 5(2-4):195–241, 2009

    2009

  23. [31]

    Greedy training algorithms for neural networks and applications to PDEs.Journal of Computational Physics, 484:112084, 2023

    Jonathan W Siegel, Qingguo Hong, Xianlin Jin, Wenrui Hao, and Jinchao Xu. Greedy training algorithms for neural networks and applications to PDEs.Journal of Computational Physics, 484:112084, 2023

    2023

  24. [32]

    Rate of convergence of two moments consistent finite volume scheme for non-classical divergence coagulation equation.Applied Numerical Mathematics, 187:120–137, 2023

    Mehakpreet Singh. Rate of convergence of two moments consistent finite volume scheme for non-classical divergence coagulation equation.Applied Numerical Mathematics, 187:120–137, 2023

    2023

  25. [33]

    Dgm: A deep learning algorithm for solving partial differential equations.Journal of Computational Physics, 375:1339–1364, 2018

    Justin Sirignano and Konstantinos Spiliopoulos. Dgm: A deep learning algorithm for solving partial differential equations.Journal of Computational Physics, 375:1339–1364, 2018

    2018

  26. [34]

    Local randomized neural networks with discontinuous Galerkin methods for partial differential equations.Journal of Computational and Applied Mathematics, 445:115830, 2024

    Jingbo Sun, Suchuan Dong, and Fei Wang. Local randomized neural networks with discontinuous Galerkin methods for partial differential equations.Journal of Computational and Applied Mathematics, 445:115830, 2024

    2024

  27. [35]

    Randomized neural networks for partial differential equation on static and evolving surfaces.arXiv preprint arXiv:2603.01689, 2026

    Jingbo Sun and Fei Wang. Randomized neural networks for partial differential equation on static and evolving surfaces.arXiv preprint arXiv:2603.01689, 2026

    2026

  28. [36]

    Numerical solution of coupled nonlinear Klein-Gordon equations on unbounded domains.Physical Review E, 106(2):025317, 2022

    Yinong Tai, Hongwei Li, Zhaojie Zhou, and Ziwen Jiang. Numerical solution of coupled nonlinear Klein-Gordon equations on unbounded domains.Physical Review E, 106(2):025317, 2022

    2022

  29. [37]

    Physics-informed neural networks combined with polynomial interpolation to solve nonlinear partial differential equations.Computers & Mathematics with Applications, 132:48–62, 2023

    Siping Tang, Xinlong Feng, Wei Wu, and Hui Xu. Physics-informed neural networks combined with polynomial interpolation to solve nonlinear partial differential equations.Computers & Mathematics with Applications, 132:48–62, 2023

    2023

  30. [38]

    Efficient spectral element method for the Euler equations on unbounded domains.Applied Mathematics and Computation, 487:129080, 2025

    Yassine Tissaoui, James F Kelly, and Simone Marras. Efficient spectral element method for the Euler equations on unbounded domains.Applied Mathematics and Computation, 487:129080, 2025

    2025

  31. [39]

    Jiangming Xie and Maojun Li. A fast BDF2 Galerkin finite element method for the one-dimensional time- dependent Schr¨ odinger equation with artificial boundary conditions.Applied Numerical Mathematics, 187:89– 106, 2023

    2023

  32. [40]

    Adaptive-distribution randomized neural networks for PDEs: A low-dimensional distribution-learning framework.arXiv preprint arXiv:2604.23999, 2026

    You Yang and Fei Wang. Adaptive-distribution randomized neural networks for PDEs: A low-dimensional distribution-learning framework.arXiv preprint arXiv:2604.23999, 2026

    2026 arXiv

  33. [41]

    Chao Zhang, Dongya Tao, and Pan Ding. Spectral method for multi-dimensional problems of high order on unbounded domains using generalized Laguerre functions.International Journal of Computer Mathematics, 98(10):2040–2059, 2021

  34. [42]

    Transferable neural networks for partial differential equations.Journal of Scientific Computing, 99(1):2, 2024

    Zezhong Zhang, Feng Bao, Lili Ju, and Guannan Zhang. Transferable neural networks for partial differential equations.Journal of Scientific Computing, 99(1):2, 2024

    2024

  35. [43]

    A highly efficient numerical method for the time-fractional diffusion equation on unbounded domains.Journal of Scientific Computing, 99(2):47, 2024

    Hongyi Zhu and Chuanju Xu. A highly efficient numerical method for the time-fractional diffusion equation on unbounded domains.Journal of Scientific Computing, 99(2):47, 2024. 41

    2024

Pith tools

Reviewed July 1, 2026 · model on record in the stance chip above.