Pith. sign in

REVIEW 4 major objections 6 minor 60 references

Advanced Physics-Informed Neural Network with Residuals for Solving Complex Integral Equations

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Adding residual skip connections to a physics-informed neural network yields lower errors on almost every integral and integro-differential equation tested, including fractional, multi-dimensional, and oscillatory-kernel problems.

desk verdict A plausible incremental idea—residual connections for PINNs solving integral equations—with a useful benchmark suite, but the paper's own tables contradict the 'consistently outperforms' claim and the experiments never isolate the residual mechanism from the PINNIES-style numerical operators. read the letter →

arxiv 2501.16370 v3 pith:3QB2EAL6 submitted 2025-01-22 cs.LG cs.AIcs.NAcs.NEmath.NA

Mahdi Movahedian Moghaddam , Kourosh Parand , Saeed Reza Kheradpisheh This is my paper · ORCID
classification cs.LGcs.AIcs.NAcs.NEmath.NA MSC 68T0765R20
keywords residualconnectionsphysics-informedneuralnetworksintegralequationsintegro-differentialfractionalderivativesGaussianquadratureoperationalmatricesHelmholtz-type
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

RISN is a physics-informed neural network for integral and integro-differential equations that adds a residual skip connection to every hidden layer, so each layer computes $\sigma(A_{i-1}\theta^{(i)}+b^{(i)}) + A_{i-1}$. The paper's claim is that this single architectural change stabilizes gradient flow and lets the network train at greater depth, producing more accurate solutions than the plain PINN and than the A-PINN and SA-PINN variants across a 20-problem benchmark suite. The measured mean absolute errors are often one to two orders of magnitude lower, and on several hard fractional, system, and oscillatory-kernel problems the variants fail to converge while RISN does. If this is right, residual connections are a cheap, broadly applicable improvement for neural solvers of nonlocal equations, not a method-specific fix.

What carries the argument

The carrying mechanism is the residual skip connection: each hidden layer is defined by $A_i = \sigma(A_{i-1}\theta^{(i)} + b^{(i)}) + A_{i-1}$, which gives gradients a direct path from the output back through the network and is credited with preventing vanishing gradients. Around this sit the numerical operators: Gauss–Legendre quadrature with 50 points evaluates integral terms, and fractional operational matrices evaluate Caputo fractional derivatives. The paper also argues from Sobolev-space approximation that the residual network learns a correction $r(x)=y^*(x)-g(x)$ to a coarse approximation, a smoother target than the raw solution, so the total approximation error inherits the network's bound on $r$.

What would settle it

Remove only the residual skip connections from RISN while keeping depth, optimizer, quadrature points, loss terms, and seeds identical, then re-run all 20 benchmarks; if the MAE does not consistently rise, the paper's central attribution of accuracy gains to residual connections is not supported.

Watch

Extended reading notes

Core claim

The central discovery claimed is that the residual connection itself—not the quadrature rule or the fractional operational matrices, which are shared with earlier PINN-based solvers—carries the accuracy gain. RISN reports the best or near-best mean absolute error in 18 of 20 benchmark problems, including a tie on the linear Abel equation, and it is the only method that converges on the nonlinear Abel problem and several first-kind systems. Across the suite the advertised improvement is up to two orders of magnitude over the baseline PINN and at least fivefold over A-PINN and SA-PINN on the hardest cases. The Helmholtz-type equation, whose solution is known only through a Neumann-series reference with grid step 0.001, is included to show that the gain persists when no exact analytical solution is available.

Load-bearing premise

The comparison assumes that the fixed 50-point Gauss–Legendre quadrature and the fractional operational matrices approximate every integral and fractional operator with error far below the network error, so that the reported MAE differences reflect the residual architecture rather than the numerical discretization.

Editorial extensions

If this is right

  • Any PINN-style integral-equation solver can adopt residual connections without changing its quadrature or loss structure, making the improvement a drop-in architectural change.
  • Deeper networks become usable: the sensitivity analysis shows smoother loss curves and lower final MAE for RISN as depth grows from 2 to 10 hidden layers, whereas PINN's MAE spikes with depth.
  • The hard cases where A-PINN and SA-PINN fail—fractional integro-differential equations, strongly coupled Volterra systems, and 2D Volterra equations—are exactly where the residual architecture matters most.
  • For singular Abel kernels, accuracy remains limited by the quadrature, not the network, so users must increase nodes or switch quadrature to push past the reported $3.27\times10^{-3}$ floor.
  • In the no-exact-solution Helmholtz problem, RISN's MAE of $1.46\times10^{-3}$ against the Neumann-series reference indicates oscillatory kernels are not a barrier to convergence, though the reference is itself a numerical construction.

Reading between the lines

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

  • Because the compared methods share the same quadrature and fractional-matrix approximations, the claimed two-order gains are best read as optimization and architecture gains for a fixed discretization; changing those discretizations could shrink or enlarge the gap depending on whether the numerical operator error dominates.
  • A direct ablation not reported in the paper—removing only the skip connections from RISN while keeping depth, optimizer, quadrature, and loss identical—would isolate whether the residual path itself is the cause of the accuracy gain.
  • The Sobolev argument suggests RISN's advantage should grow when a good coarse approximation $g$ is available; testing the method with deliberately poor choices of $g$ would separate the benefit of a smoother correction target from the benefit of improved gradient flow.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes RISN, a fully connected neural network with residual (skip) connections for solving integral and integro-differential equations, including one- and multi-dimensional problems, systems, fractional equations, and a Helmholtz-type oscillatory-kernel problem. Integral terms are evaluated with Gauss-Legendre quadrature and fractional derivatives with operational matrices, and training minimizes a weighted MSE of the equation residual plus initial/boundary/data terms. The paper compares RISN with PINN, A-PINN, and SA-PINN on 20 benchmark problems, reporting MAE values, and claims that residual connections stabilize gradient flow and yield consistently lower MAE than all baselines.

Significance. If substantiated, a robust architecture covering this breadth of integral and integro-differential equations would be a useful contribution to scientific machine learning. The paper has real strengths: the benchmark suite is broad, MAE is measured against independent exact solutions for most problems, the experimental section states that all methods share the same network architecture and optimizer, and the sensitivity analysis addresses depth and learning rate. However, the central quantitative claim is not supported by the paper's own data, and the attribution of the gains to residual connections is confounded by the numerical-operator implementation. The result is potentially salvageable, but only after a controlled comparison and a substantially revised set of claims.

major comments (4)
  1. [Section 4.3, Table 6] The abstract and Sections 5-6 claim that RISN 'consistently outperforms' baselines and 'achieves the lowest MAE across almost all problems,' but Table 6 contradicts this. Problem 15 (first-kind Volterra ordinary integro-differential equation) reports A-PINN MAE 1.70e-5 versus RISN MAE 5.38e-4, so A-PINN is roughly 30 times more accurate. In Problem 10, second component, PINN achieves 1.98e-5 versus RISN 1.18e-4, and in Problem 5 the values are an exact tie at 3.27e-3. Table 3 already concedes one Volterra case where PINN outperforms RISN. The overarching performance claim must be replaced by a per-problem, quantified statement, and the distribution of wins/losses/tie should be reported explicitly.
  2. [Sections 3.2 and 4] The claimed causal role of residual connections is not isolated. Section 3.2 credits PINNIES [39] with first integrating Gaussian quadrature and fractional operational matrices into a PINN framework and states that these techniques 'do not significantly differ from their original implementation.' The experimental protocol in Section 4 guarantees identical network architecture and optimizer settings across methods, but it never states that the PINN, A-PINN, and SA-PINN baselines evaluate integral and fractional operators with the same 50-point Gauss-Legendre rule and the same fractional operational matrices. A residual-free PINNIES baseline with identical numerical operators is the natural control and is absent. Because Section 4.1.1 itself notes that Abel-type accuracy depends on the number of quadrature nodes or the choice of quadrature rule, the numerical layer can dominate the error; without this control, the improvements in Tables 1-6 cannot be attributed to residual connections.
  3. [Section 3.3] The Sobolev-space argument is not a proof of an advantage for RISN. It assumes a known approximation g with small W^{k,p} error and asserts that the residual r is smoother, but no such g is constructed in the RISN pipeline, no approximation rate with explicit dependence on width and depth is stated, and the cited results [58-60] are universal-approximation statements rather than quantitative Sobolev bounds for the specific residual decomposition used here. The 'implication' therefore does not follow from the cited mathematics. This subsection should either be removed or replaced with a rigorous statement; in its current form it overstates the theoretical grounding.
  4. [Tables 1-6 and Figure 3] All experimental comparisons report single MAE values with no number of seeds, repeated runs, error bars, or statistical tests. Since L-BFGS training is initialization-dependent, claims of 'robustness,' 'stability,' and 'consistently' lower error are not supported by the evidence. The absence of code or data also prevents checking whether the baseline implementations share the numerical operators. Reporting at least a small number of independent runs with dispersion measures is necessary before the comparative conclusions can be drawn.
minor comments (6)
  1. [Equation (3)] The loss definition repeats lambda_IC; the second coefficient should presumably be lambda_BC.
  2. [Section 3.1 and Figure 1] The roles of F, D, and I are described inconsistently: one sentence says F(u), D(u), I(u) correspond to differential, integral, and source terms, while the following text says D(u) is the differential operator. Clarify the notation.
  3. [Section 4.4] The Helmholtz problem uses a Neumann-series reference solution on a discretization grid of 0.001, but no validation of the reference accuracy is given. Without an error estimate for the reference, statements such as 'MAE of 1.46e-3' versus '7.51e-3' assume that the reference is much more accurate than both values.
  4. [Introduction and Section 3.2] The paper lists 'integration of advanced numerical techniques' as a contribution, but Section 3.2 states that the Gaussian quadrature and fractional operational matrices were first integrated into a PINN framework by PINNIES [39] and 'do not significantly differ from their original implementation.' The novelty framing should be revised to focus on the residual architecture and to acknowledge that the numerical techniques are inherited.
  5. [Section 4.3.1] The bullet 'In 18 out of 20 problems, RISN delivers either the best or near-best accuracy' is unverifiable because 'near-best' is not defined. Provide an exact ranking criterion and a complete win/loss/tie table.
  6. [Tables 1-6] Several table headings contain typos such as 'T able' and 'F ractional'; these should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: RISN's accuracy claims are evaluated against independent exact or reference solutions, and the PINNIES self-citation covers standard numerical operators rather than defining the residual-connection result.

full rationale

The paper's central claim is that adding residual connections to a PINN-style integral-equation solver improves accuracy and stability. That claim is tested by computing MAE against known exact solutions (and, for the Helmholtz problem, an independently constructed Neumann-series reference), not by fitting or renaming the outcome. The only overlapping-author citation is PINNIES [39], credited for Gaussian quadrature and fractional operational matrices. Those techniques are standard numerical tools, and the paper explicitly says they 'do not significantly differ from their original implementation'; they are inputs shared with the numerical layer, not definitions of the residual architecture's predicted benefit. No equation is shown to equal an input by construction, no fitted parameter is relabeled as a prediction, and no load-bearing argument reduces to an unverified self-citation. The absence of a PINNIES-without-residuals control is a legitimate experimental-confound concern, but it is not circularity under the defined criteria.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The architecture uses standard hyperparameters; the most consequential assumptions are the accuracy of the quadrature and operational matrix discretization and the fairness of the optimizer in comparing baseline methods.

free parameters (5)
  • Loss weights lambda_IC, lambda_BC, lambda_Data = set equal (unweighted sum)
    Eq. (3) defines them but Section 4 states all components contribute equally; no values or tuning reported.
  • Network depth and width = 7 hidden layers x 20 neurons
    Chosen by hand; sensitivity to depth is tested only for PINN vs RISN, not for other baselines.
  • Number of training points N = 50
    Set to 50 for all problems; no convergence study in N.
  • Gauss-Legendre quadrature order = 50
    Fixed at 50; the paper notes Abel problems need more nodes or a different quadrature, so this choice affects reported MAE.
  • Optimizer learning rate = 0.01
    Set for L-BFGS; sensitivity analysis compares 1e-2 and 1e-3, but baselines use the same setting.
assumptions (5)
  • standard math Universal approximation theorems and Sobolev approximation rates for neural networks
    Section 3.3 invokes [58-60] to claim the residual correction network can approximate r(x) to arbitrary accuracy.
  • domain assumption Gaussian quadrature converges for the kernel and source integrands in all benchmark problems
    Section 3.2.1 assumes quadrature error is negligible; Section 4.1.1 concedes this fails for Abel singular kernels.
  • domain assumption The fractional operational matrix from PINNIES computes the Caputo derivative correctly for the trial solution space
    Section 3.2.2 relies on the matrix representation without derivation or validation; the fractional example's improvement is small (3.64e-3 vs 5.50e-3).
  • domain assumption The benchmark source terms and exact solutions are correctly constructed and are the correct solutions of the stated equations
    Section 4 defines exact solutions for each problem; any algebra error would bias MAE. No independent verification.
  • ad hoc to paper L-BFGS optimization reaches a sufficiently good local minimum for all methods
    The comparison assumes failures in '-' entries are due to model limitations rather than optimization or tuning artifacts.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Advanced Physics-Informed Neural Network with Residuals for Solving Complex Integral Equations." pith.science (2026). https://pith.science/paper/3QB2EAL6

@misc{pith2026250116370,
  author       = {Pith},
  title        = {Pith review of: Advanced Physics-Informed Neural Network with Residuals for Solving Complex Integral Equations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3QB2EAL6}},
  note         = {Machine review of arXiv:2501.16370}
}
read the original abstract

In this paper, we present the Residual Integral Solver Network (RISN), a novel neural network architecture designed to solve a wide range of integral and integro-differential equations, including one-dimensional, multi-dimensional, ordinary and partial integro-differential, systems, fractional types, and Helmholtz-type integral equations involving oscillatory kernels. RISN integrates residual connections with high-accuracy numerical methods such as Gaussian quadrature and fractional derivative operational matrices, enabling it to achieve higher accuracy and stability than traditional Physics-Informed Neural Networks (PINN). The residual connections help mitigate vanishing gradient issues, allowing RISN to handle deeper networks and more complex kernels, particularly in multi-dimensional problems. Through extensive experiments, we demonstrate that RISN consistently outperforms not only classical PINNs but also advanced variants such as Auxiliary PINN (A-PINN) and Self-Adaptive PINN (SA-PINN), achieving significantly lower Mean Absolute Errors (MAE) across various types of equations. These results highlight RISN's robustness and efficiency in solving challenging integral and integro-differential problems, making it a valuable tool for real-world applications where traditional methods often struggle.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 51 canonical work pages

  1. [39]

    PINNIES: An Efficient Physics-Informed Neural Network Framework to Integral Operator Problems

    Aghaei, A.A., Moghaddam, M.M., Parand, K.: PINNIES: An Efficient Physics- Informed Neural Network Framework to Integral Operator Problems (2024). https://arxiv.org/abs/2409.01899

    work page Pith review arXiv 2024

  2. [1]

    Wazwaz, A.-M.: Linear and Nonlinear Integral Equations vol. 639. Springer, ??? (2011)

    work page 2011

  3. [2]

    Springer (1989)

    Kress, R.: Linear integral equations. Springer (1989)

    work page 1989

  4. [3]

    SIAM review39(3), 484–493 (1997)

    TeBeest, K.G.: Classroom note: numerical and analytical solutions of volterra’s population model. SIAM review39(3), 484–493 (1997)

    work page 1997

  5. [4]

    Proceedings of the royal society of london

    Kermack, W.O., McKendrick, A.G.: A contribution to the mathematical theory of epidemics. Proceedings of the royal society of london. Series A, Containing papers of a mathematical and physical character115(772), 700–721 (1927)

    1927

  6. [5]

    Springer, ??? (2012)

    Estrada, R., Kanwal, R.P.: Singular Integral Equations. Springer, ??? (2012)

    work page 2012

  7. [6]

    Theoretical population biology 2(1), 1–23 (1971)

    Scudo, F.M.: Vito volterra and theoretical ecology. Theoretical population biology 2(1), 1–23 (1971)

    work page 1971

  8. [7]

    Thermochimica Acta660, 82–93 (2018) 24

    Minakov, A.A., Schick, C.: Non-equilibrium fast thermal response of polymers. Thermochimica Acta660, 82–93 (2018) 24

    work page 2018

Show all 60 references
  1. [8]

    International journal of computer mathematics61(1-2), 103–110 (1996)

    Khuri, S.A., Wazwaz, A.-M.: The decomposition method for solving a second kind fredholm integral equation with a logarithmic kernel. International journal of computer mathematics61(1-2), 103–110 (1996)

    1996

  2. [9]

    Thomas, J.W.: Numerical Partial Differential Equations: Finite Difference Meth- ods vol. 22. Springer, ??? (2013)

    2013

  3. [10]

    Mathematics and Computers in Simulation71(1), 16–30 (2006)

    Dehghan, M.: Finite difference procedures for solving a problem arising in mod- eling and design of certain optoelectronic devices. Mathematics and Computers in Simulation71(1), 16–30 (2006)

    2006

  4. [11]

    Elsevier, ??? (2005)

    Zienkiewicz, O.C., Taylor, R.L.: The Finite Element Method for Solid and Structural Mechanics. Elsevier, ??? (2005)

    2005

  5. [12]

    Polymers 12(4), 818 (2020)

    David M¨ uzel, S., Bonhin, E.P., Guimar˜ aes, N.M., Guidi, E.S.: Application of the finite element method in the analysis of composite materials: A review. Polymers 12(4), 818 (2020)

    2020

  6. [13]

    Courier Corporation, ??? (2001)

    Boyd, J.P.: Chebyshev and Fourier Spectral Methods. Courier Corporation, ??? (2001)

    2001

  7. [14]

    Engineering with Computers38(3), 1977–1984 (2022)

    Yousefi, F.S., Ordokhani, Y., Yousefi, S.: Numerical solution of variable order fractional differential equations by using shifted legendre cardinal functions and ritz method. Engineering with Computers38(3), 1977–1984 (2022)

    2022

  8. [15]

    The European Physical Journal Plus135, 1–19 (2020)

    Latifi, S., Parand, K., Delkhosh, M.: Generalized lagrange–jacobi–gauss–radau collocation method for solving a nonlinear optimal control problem with the classical diffusion equation. The European Physical Journal Plus135, 1–19 (2020)

    2020

  9. [16]

    Journal of Complexity26(5), 455–489 (2010)

    Griebel, M., Holtz, M.: Dimension-wise integration of high-dimensional functions with applications to finance. Journal of Complexity26(5), 455–489 (2010)

    2010

  10. [17]

    part ii: implementation and tests in dune

    Bastian, P., Blatt, M., Dedner, A., Engwer, C., Kl¨ ofkorn, R., Kornhuber, R., Ohlberger, M., Sander, O.: A generic grid interface for parallel and adaptive scientific computing. part ii: implementation and tests in dune. Computing82, 121–138 (2008)

    2008

  11. [18]

    SIAM, ??? (2002)

    Ciarlet, P.G.: The Finite Element Method for Elliptic Problems. SIAM, ??? (2002)

    2002

  12. [19]

    In: SC’12: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, pp

    Sundar, H., Biros, G., Burstedde, C., Rudi, J., Ghattas, O., Stadler, G.: Par- allel geometric-algebraic multigrid on unstructured forests of octrees. In: SC’12: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, pp. 1–...

    2012

  13. [20]

    Springer, ??? (2011)

    Tarasov, V.E.: Fractional Dynamics: Applications of Fractional Calculus to 25 Dynamics of Particles, Fields and Media. Springer, ??? (2011)

    2011

  14. [21]

    John wiley & sons, ??? (1991)

    Atkinson, K.: An Introduction to Numerical Analysis. John wiley & sons, ??? (1991)

    1991

  15. [22]

    Journal of Vibration and Control17(13), 2050– 2058 (2011)

    Saadatmandi, A., Dehghan, M.: A legendre collocation method for fractional integro-differential equations. Journal of Vibration and Control17(13), 2050– 2058 (2011)

    2011

  16. [23]

    MIT press (2016)

    Goodfellow, I.: Deep learning. MIT press (2016)

    2016

  17. [24]

    nature521(7553), 436–444 (2015)

    LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. nature521(7553), 436–444 (2015)

    2015

  18. [25]

    Journal of Computational physics378, 686–707 (2019)

    Raissi, M., Perdikaris, P., Karniadakis, G.E.: Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational physics378, 686–707 (2019)

    2019

  19. [26]

    Nature Reviews Physics3(6), 422–440 (2021)

    Karniadakis, G.E., Kevrekidis, I.G., Lu, L., Perdikaris, P., Wang, S., Yang, L.: Physics-informed machine learning. Nature Reviews Physics3(6), 422–440 (2021)

    2021

  20. [27]

    SIAM review63(1), 208–228 (2021)

    Lu, L., Meng, X., Mao, Z., Karniadakis, G.E.: Deepxde: A deep learning library for solving differential equations. SIAM review63(1), 208–228 (2021)

    2021

  21. [28]

    Journal of Computational Physics357, 125–141 (2018)

    Raissi, M., Karniadakis, G.E.: Hidden physics models: Machine learning of nonlin- ear partial differential equations. Journal of Computational Physics357, 125–141 (2018)

    2018

  22. [29]

    University of Notre Dame, ??? (2019)

    Zhu, Y.: Data-Driven and Physics-Constrained Deep Learning for Surrogate Mod- eling and Uncertainty Quantification of Physical Systems. University of Notre Dame, ??? (2019)

    2019

  23. [30]

    Neurocomputing317, 28–41 (2018)

    Berg, J., Nystr¨ om, K.: A unified deep artificial neural network approach to partial differential equations in complex geometries. Neurocomputing317, 28–41 (2018)

    2018

  24. [31]

    Journal of computational physics375, 1339–1364 (2018)

    Sirignano, J., Spiliopoulos, K.: Dgm: A deep learning algorithm for solving partial differential equations. Journal of computational physics375, 1339–1364 (2018)

    2018

  25. [32]

    Proceedings of the National Academy of Sciences115(34), 8505–8510 (2018)

    Han, J., Jentzen, A., E, W.: Solving high-dimensional partial differential equations using deep learning. Proceedings of the National Academy of Sciences115(34), 8505–8510 (2018)

    2018

  26. [33]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recogni- tion. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016) 26

    2016

  27. [34]

    IEEE transactions on neural networks5(2), 157–166 (1994)

    Bengio, Y., Simard, P., Frasconi, P.: Learning long-term dependencies with gra- dient descent is difficult. IEEE transactions on neural networks5(2), 157–166 (1994)

    1994

  28. [35]

    International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems6(02), 107–116 (1998)

    Hochreiter, S.: The vanishing gradient problem during learning recurrent neural nets and problem solutions. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems6(02), 107–116 (1998)

    1998

  29. [36]

    arXiv preprint arXiv:2308.05423 (2023)

    Gazoulis, D., Gkanis, I., Makridakis, C.G.: On the stability and convergence of physics informed neural networks. arXiv preprint arXiv:2308.05423 (2023)

    2023 arXiv

  30. [37]

    Journal of Computational Physics462, 111260 (2022)

    Yuan, L., Ni, Y.-Q., Deng, X.-Y., Hao, S.: A-pinn: Auxiliary physics informed neural networks for forward and inverse problems of nonlinear integro-differential equations. Journal of Computational Physics462, 111260 (2022)

    2022

  31. [38]

    Journal of Computational Physics474, 111722 (2023)

    McClenny, L.D., Braga-Neto, U.M.: Self-adaptive physics-informed neural net- works. Journal of Computational Physics474, 111722 (2023)

    2023

  32. [40]

    IEEE Transactions on Antennas and Propagation56(8), 2442–2452 (2008)

    Wang, X., Wildman, R.A., Weile, D.S., Monk, P.: A finite difference delay modeling approach to the discretization of the time domain integral equations of electromagnetics. IEEE Transactions on Antennas and Propagation56(8), 2442–2452 (2008)

    2008

  33. [41]

    Superconductor Science and Technology21(10), 105008 (2008)

    Brambilla, R., Grilli, F., Martini, L., Sirois, F.: Integral equations for the cur- rent density in thin conductors and their solution by the finite-element method. Superconductor Science and Technology21(10), 105008 (2008)

    2008

  34. [42]

    Springer, ??? (2024)

    Vandandoo, U., Zhanlav, T., Chuluunbaatar, O., Gusev, A., Vinitsky, S., Chu- luunbaatar, G.: High-Order Finite Difference and Finite Element Methods for Solving Some Partial Differential Equations. Springer, ??? (2024)

    2024

  35. [43]

    Journal of Scientific Computing 90(1), 35 (2021) https://doi.org/10.1007/s10915-021-01714-6

    Hamfeldt, B.F., Lesniewski, J.: Convergent finite difference methods for fully nonlinear elliptic equations in three dimensions. Journal of Scientific Computing 90(1), 35 (2021) https://doi.org/10.1007/s10915-021-01714-6

    2021 doi

  36. [44]

    Earth, Planets and Space70(1), 103 (2018) https://doi.org/10.1186/ s40623-018-0876-7

    Sarakorn, W., Vachiratienchai, C.: Hybrid finite difference–finite element method to incorporate topography and bathymetry for two-dimensional magnetotelluric modeling. Earth, Planets and Space70(1), 103 (2018) https://doi.org/10.1186/ s40623-018-0876-7

    2018

  37. [45]

    Archives of Computational Methods in Engineering29(6), 4431–4453 (2022) https://doi.org/10.1007/s11831-022-09740-9 27

    Liu, W.K., Li, S., Park, H.S.: Eighty years of the finite element method: Birth, evolution, and future. Archives of Computational Methods in Engineering29(6), 4431–4453 (2022) https://doi.org/10.1007/s11831-022-09740-9 27

    2022 doi

  38. [46]

    Archives of Computational Methods in Engineering31(5), 2945–2984 (2024) https://doi.org/ 10.1007/s11831-024-10063-0

    Nath, D., Ankit, Neog, D.R., Gautam, S.S.: Application of machine learning and deep learning in finite element analysis: A comprehensive review. Archives of Computational Methods in Engineering31(5), 2945–2984 (2024) https://doi.org/ 10.1007/s11831-024-10063-0

    2024 doi

  39. [47]

    Mathematics 11(21) (2023) https://doi.org/10.3390/math11214418

    Chen, X., Zhang, K., Ji, Z., Shen, X., Liu, P., Zhang, L., Wang, J., Yao, J.: Progress and challenges of integrated machine learning and traditional numerical algorithms: Taking reservoir numerical simulation as an example. Mathematics 11(21) (2023) https://doi.org/10.3390/mat...

    2023 doi

  40. [48]

    Neural Computing and Applications21, 843–852 (2012)

    Effati, S., Buzhabadi, R.: A neural network approach for solving fredholm integral equations of the second kind. Neural Computing and Applications21, 843–852 (2012)

    2012

  41. [49]

    Iranian journal of fuzzy systems11(1), 91–112 (2014)

    Mosleh, M.: Numerical solution of fuzzy linear fredholm integro-differential equation by\\fuzzy neural network. Iranian journal of fuzzy systems11(1), 91–112 (2014)

    2014

  42. [50]

    Applied Soft Computing 27, 391–398 (2015)

    Jafarian, A., Measoomy, S., Abbasbandy, S.: Artificial neural networks based modeling for solving volterra integral equations system. Applied Soft Computing 27, 391–398 (2015)

    2015

  43. [51]

    Mathematical Sciences 8, 1–9 (2014)

    Asady, B., Hakimzadegan, F., Nazarlue, R.: Utilizing artificial neural network approach for solving two-dimensional integral equations. Mathematical Sciences 8, 1–9 (2014)

    2014

  44. [52]

    Chaharborj, S.S., Chaharborj, S., Mahmoudi, Y.: Study of fractional order integro-differential equations by using chebyshev neural network. J. Math. Stat 13(1), 1–13 (2017)

    2017

  45. [53]

    IEEE Transactions on Antennas and Propagation70(8), 6135–6147 (2021)

    Guo, R., Shan, T., Song, X., Li, M., Yang, F., Xu, S., Abubakar, A.: Physics embedded deep neural network for solving volume integral equation: 2-d case. IEEE Transactions on Antennas and Propagation70(8), 6135–6147 (2021)

    2021

  46. [54]

    Nature Machine Intelligence3(3), 218–229 (2021) https://doi.org/10.1038/ s42256-021-00302-5

    Lu, L., Jin, P., Pang, G., Zhang, Z., Karniadakis, G.E.: Learning nonlinear operators via deeponet based on the universal approximation theorem of opera- tors. Nature Machine Intelligence3(3), 218–229 (2021) https://doi.org/10.1038/ s42256-021-00302-5

    2021

  47. [55]

    https://arxiv.org/abs/2010.08895

    Li, Z., Kovachki, N., Azizzadenesheli, K., Liu, B., Bhattacharya, K., Stuart, A., Anandkumar, A.: Fourier Neural Operator for Parametric Partial Differential Equations (2021). https://arxiv.org/abs/2010.08895

    2021 arXiv

  48. [56]

    Martire, A.L., Congedo, M.A., Cenci, M.: Fractional Volterra Integral Equations: A Neural Network Approach vol. 1. Roma TrE-Press, ??? (2022) 28

    2022

  49. [57]

    Information Sciences612, 887–897 (2022) https: //doi.org/10.1016/j.ins.2022.09.017

    Saneifard, R., Jafarian, A., Ghalami, N., Nia, S.M.: Extended artificial neu- ral networks approach for solving two-dimensional fractional-order volterra-type integro-differential equations. Information Sciences612, 887–897 (2022) https: //doi.org/10.1016/j.ins.2022.09.017

    2022 doi

  50. [58]

    Neural networks3(5), 551–560 (1990)

    Hornik, K., Stinchcombe, M., White, H.: Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks. Neural networks3(5), 551–560 (1990)

    1990

  51. [59]

    Acta numerica8, 143–195 (1999)

    Pinkus, A.: Approximation theory of the mlp model in neural networks. Acta numerica8, 143–195 (1999)

    1999

  52. [60]

    Neural computation8(1), 164–177 (1996) 29

    Mhaskar, H.N.: Neural networks for optimal approximation of smooth and analytic functions. Neural computation8(1), 164–177 (1996) 29

    1996

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.