Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

Neural Preconditioning Operator for Efficient PDE Solves

T0 review · 3 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read The NPO method learns a multigrid-style preconditioner that accelerates Krylov solvers for elliptic PDEs and transfers to grids far beyond its training resolution.

desk verdict The central claim that NPO beats traditional solvers is untested because AMG, the solver used to generate the training data, never appears as a baseline; the appendix also contains physically impossible memory numbers. read the letter →

arxiv 2502.01337 v2 pith:3SGRJHHK submitted 2025-02-03 cs.CE

Zhihao Li , Di Xiao , Zhilu Lai , Wei Wang This is my paper · ORCID
classification cs.CE MSC 65F1065N2268T07
keywords neuralpreconditioningalgebraicmultigridoperatorsKrylovsubspacemethodsPDEsolverstransformerattentioneigenvalueclusteringresolutiongeneralization
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

This paper proposes a way to speed up solving discretized PDEs: train a neural network to act as a preconditioner for Krylov iterative solvers, the class of linear-system solvers that build approximations in nested subspaces. The network, called the Neural Preconditioning Operator (NPO), learns a matrix $M$ that approximates the inverse of the system matrix $A$, so that the preconditioned problem $M A x = M b$ has eigenvalues clustered near 1 and converges in fewer iterations. The architecture blends algebraic multigrid with a transformer-based attention mechanism, so it reduces both high- and low-frequency errors on coarse and fine grids. The paper reports that on Poisson, diffusion, and linear elasticity problems, NPO converges in fewer iterations and less wall-clock time than classical and neural baselines, and that a model trained on a 128 resolution still works on grids up to 4096.

What carries the argument

The load-bearing object is the NAMG (Neural Algebraic Multigrid) operator, which replaces the hand-built restriction and prolongation matrices of algebraic multigrid with learned attention weights. In the paper's notation, restriction is $R = A E$ and prolongation is $P = A E^T$, where $E$ contains softmax attention scores over graph neighborhoods; coarse features are refined by self-attention, then corrected back to the fine grid. The training objectives are the condition loss, which penalizes the norm of $(I - A M(A)) r$ over sampled residuals, and the residual loss, which penalizes the norm of $A M(A) b - b$; these steer $M$ toward $A^{-1}$ without forming the full matrix inverse. The theoretical part of the paper uses the standard two-grid contraction argument, whose contraction factor depends on the Approximation Property and smoothing stability, plus a spectral argument that $M A$ has eigenvalues in an interval near 1.

What would settle it

Run the trained NAMG as a two-grid cycle on a Poisson problem at several grid sizes, starting from a random error, and measure the energy-norm contraction ratio per cycle and the largest eigenvalue spread of the preconditioned matrix $M A$; if the measured contraction ratio is not below 1 or becomes close to 1 as n grows, or if the eigenvalues of $M A$ are not clustered near 1, then the claimed convergence and spectral guarantees fail.

Watch

Extended reading notes

Core claim

The central claim is that a data-driven operator trained with condition, residual, and data losses can serve as an effective preconditioner for Krylov solvers across mesh types and PDE families. The learned operator is explicitly constructed as a Neural Algebraic Multigrid (NAMG), with learned attention-based restriction and prolongation, and the paper argues that this construction inherits the two-grid convergence guarantee and the eigenvalue-clustering property of classical multigrid, so that the preconditioned matrix $M A$ behaves close to the identity. The numerical evidence is that NPO reaches a relative residual of 1e-10 on a 512-grid Poisson problem in 184 iterations and 0.623 seconds versus 513 iterations and 3.719 seconds for Jacobi, and that it remains competitive on irregular meshes and on grids up to 4096 although training was done at 128.

Load-bearing premise

The proof that NPO converges relies on the assumption, never verified numerically, that the learned attention-based restriction and prolongation operators satisfy the multigrid Approximation Property and smooth stably; if the trained network fails these conditions, the two-grid contraction guarantee does not apply to the actual learned preconditioner.

Editorial extensions

If this is right

  • On the tested elliptic problems (Poisson, diffusion, linear elasticity), the preconditioned GMRES converges in fewer iterations and less time than Jacobi, Gauss–Seidel, SOR, and the listed neural baselines, at tolerances from 1e-4 to 1e-10.
  • A model trained only at resolution 128 retains useful preconditioning quality at 1024, 2048, and 4096, so the mesh resolution does not need to be a training parameter.
  • The theoretical analysis, conditional on the Approximation Property and smoothing stability of the learned operators, implies the two-grid iteration contracts the error in the energy norm at a rate independent of system size, and that the preconditioned system has a condition number near 1.
  • Ablations show that the NAMG structure, the system matrix $A$ as input, and the data loss each contribute materially; removing the full NAMG raises iteration count from 184 to 314, and removing the data loss from 184 to 206.

Reading between the lines

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

  • If the spectral clustering claim holds for trained operators across PDE families, the same learned $M$ could be reused as a warm-start preconditioner inside nonlinear solves or time-stepping schemes that repeatedly solve nearby linear systems, where the reported generalization is most valuable.
  • The reported parameter count (0.14 MB) suggests NPO captures structural features of the discrete operator rather than memorizing a particular mesh or right-hand side; a test to confirm this is to train once on one domain shape and evaluate on a topologically different mesh with holes or reentrant corners.
  • Because the condition and residual losses are defined only through matrix-vector products with $A$, the framework could extend to implicit or matrix-free operators where $A$ is never explicitly assembled, broadening its reach beyond the symmetric positive-definite elliptic cases shown.
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

3 major / 6 minor

Summary. The paper introduces the Neural Preconditioning Operator (NPO), a learned preconditioner that combines an algebraic-multigrid-inspired restriction/prolongation structure with transformer-style attention and is trained with data, residual, and condition losses. The learned operator is applied as a left preconditioner for Krylov solvers (GMRES/CG). The authors claim two-grid convergence and spectrum-clustering guarantees for the NAMG operator, and report experiments on Poisson, diffusion, and linear elasticity problems on uniform and irregular meshes, showing lower iteration counts and runtimes than Jacobi, Gauss–Seidel, SOR, MLP, U-Net, FNO, Transolver, and M2NO, plus a resolution-generalization study up to 4096.

Significance. If the empirical comparison were complete, the reported result that NPO converges to 1e-10 on the 512 Poisson problem in 184 iterations and 0.623 s versus 513 iterations and 3.719 s for Jacobi would be a useful demonstration of a learned preconditioner for small elliptic systems. The paper is also valuable for articulating concrete losses (condition and residual) for training preconditioners and for combining multigrid ideas with neural attention. However, the significance is currently limited by two load-bearing issues: (i) the training data are generated with an AMG-preconditioned Krylov solver that is never included as a baseline, so the central claim of outperforming traditional methods is not tested against the most relevant classical competitor; and (ii) the theoretical guarantees in Section 4 are conditional on unverified multigrid properties of the learned operators. No code or data are provided, which also makes reproducibility harder to assess.

major comments (3)
  1. [§5.1.2, Tables 1–2, Figure 4] The dataset generation (Section 5.1.2) states that each system is solved with a baseline Krylov solver (CG or GMRES) preconditioned by Algebraic Multigrid (AMG) with tolerance 1e-10 and a cap of 100 iterations, and the recorded partial solutions and residuals are used to train M_theta. Yet AMG is absent from every experimental comparison: Tables 1 and 2 and Figure 4 compare NPO only against Jacobi, Gauss–Seidel, SOR, and neural baselines. Since NPO is trained to imitate the AMG-preconditioned solver's behavior, omitting that solver removes the most relevant classical competitor. Please add AMG-preconditioned CG/GMRES as a baseline in all experiments and report both iteration counts and wall-clock times. If AMG already converges in tens of iterations on the 512 Poisson problem, then NPO's 184 iterations would not support the abstract's claim that NPO 'significantly reduces iteration counts' relative to traditional methods. This is the load-bearing issue for the paper's primary empirical claim.
  2. [§4, Appendix C.2–C.3] The convergence guarantee in Theorem 4.1 and the spectrum-clustering guarantee in Theorem 4.3 are derived under the Approximation Property (Property 4.2) and smoothing stability for the restriction, prolongation, and smoothing operators. For the neural setting, these properties are assumed, not verified for the trained NPO. The proof of Theorem 4.3 in Appendix C.3 is a qualitative restatement of classical multigrid theory and does not establish quantitative eigenvalue bounds for the learned operator. Please either (a) verify the assumptions numerically for the trained networks, e.g., by estimating the approximation constant alpha and the smoothing factor nu on test matrices, or (b) state explicitly that the theoretical results apply only if the learned operators satisfy these properties. As written, the conclusion in Section 6 that 'theoretical analysis confirms that NAMG inherits two-grid convergence guarantees' is stronger than what is proven.
  3. [§5.3, Figure 4] The resolution-generalization claim—that NPO maintains robust convergence on grids as large as 4096—is supported only by an unlabeled plot. Figure 4 has no numeric values, so the reader cannot check the claimed iteration counts or the relative scaling of NPO versus the other methods. Please provide a table or labeled points with exact iteration counts (and ideally wall-clock times) for every method at resolutions 128, 1024, 2048, and 4096, including the training resolution 128. Note also that Table 5 reports training-phase efficiency only; it does not quantify solve-time scaling.
minor comments (6)
  1. [Appendix A] Appendix A contains unresolved reference placeholders '[? ? ]' and '[? ]' in the matrix reordering and multilevel methods discussion, and the section title 'Neural Preconditoner' has a typo ('Preconditoner' should be 'Preconditioner').
  2. [Figure 3, Table 1, §5.2] The text in Section 5.2 and the Figure 3 caption refer to a '512 × 512' grid, while Table 1 and the surrounding text describe the experiment as 'Grid=512' and '512 grid'. Please clarify whether the Poisson experiment is one-dimensional or two-dimensional, since this affects the interpretation of iteration counts and runtimes.
  3. [Table 1] Table 1 has duplicated and misaligned column headers (e.g., 'ITERATION' and '1E-10' appear twice under each grid type). Reformat the table so each tolerance column is clearly labeled once.
  4. [§5.1.1, §3.3, Table 4] The training objective in Section 5.1.1 mentions only Data Loss and Residual Loss, but Section 3.3 defines a Condition Loss as well. It is unclear how the three losses are combined and weighted; the hyperparameter study in Table 4 also does not report loss-weight settings. Please state the full combined objective and the loss-weight values used.
  5. [Appendix D.2] The heading 'Primary configurations for Nerual Preconditioning Operator' contains a typo ('Nerual' should be 'Neural').
  6. [Theorem 4.4, Appendix C.4] Theorem 4.4 and its proof show that the attention aggregation can be viewed as a Monte-Carlo approximation of an integral operator. This is a representational observation rather than a quantitative convergence guarantee; labeling it a theorem may overstate its role in the paper's theoretical contributions.

Circularity Check

0 steps flagged · score 2.0 of 10

No definitional circularity; the main empirical and theoretical claims rest on external benchmarks and conditional theorems, with the missing AMG baseline an evaluation gap rather than an input-output reduction.

full rationale

No load-bearing step in the derivation chain reduces to its own inputs by the paper's own equations. The condition and residual losses (Eqs. 9-10) deliberately train M_theta to approximate A^{-1}, so the later report of reduced GMRES iteration counts is the expected consequence of optimizing that objective on the same PDE family; this is a fitted empirical outcome, not a parameter-free prediction, but it is not circular because the GMRES iteration count is an externally measured quantity rather than a term in the loss. The use of AMG-preconditioned Krylov trajectories in Section 5.1.2 to generate training residuals, combined with the absence of an AMG baseline in Tables 1-2, is a serious benchmarking and validity concern, but it is an omitted-control problem, not a definitional or self-referential reduction. The theoretical results (Theorems 4.1 and 4.3) are explicitly conditional on the standard multigrid Approximation Property and smoothing stability; the paper does not verify these properties for the learned operators, so the convergence guarantee is an assumption gap, not a circular derivation. Self-citations ([16] M2NO as baseline, [17] for the graph-attention-as-integral equivalence in Appendix C.4) are peripheral and not load-bearing for the central claim. Under the evidentiary bar of exhibiting a specific equation-level reduction, no circular step is established.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. The NPO and NAMG are network architectures, not postulated physical objects. The free parameters are standard ML hyperparameters and an unreported loss-weighting scheme.

free parameters (5)
  • feature_width = 32
    Chosen by hyperparameter sweep on Poisson (Table 4). Values of 64 and 128 increase iteration counts to 308 and 356.
  • num_c (number of coarse points) = 128
    Chosen by sweep; using 8 or 64 coarse points leads to over 200 iterations.
  • num_heads = 4
    Chosen by sweep; 8 or 16 heads increase complexity without reducing iteration counts.
  • pre_ite and post_ite = 1
    A single pre/post relaxation pass is optimal in the sweep; more passes increase iteration counts.
  • loss weighting = not reported
    The three losses (data, residual, condition) are summed without stated coefficients. Any implicit weighting is a hand-chosen free parameter.
assumptions (4)
  • standard math Classical two-grid convergence theory: if smoothing and approximation properties hold, the iteration contracts by a constant rho < 1 independent of n.
    Invoked in Theorem 4.1 and proof in Appendix C.2. This is standard multigrid theory from reference [2].
  • ad hoc to paper The learned restriction and prolongation operators P and R satisfy the Approximation Property (Property 4.2) and smoothing stability.
    Never proved for the NAMG architecture. The proofs in Appendix C.2 and C.3 assume the existence of smoothing factor nu and approximation factor alpha for the neural operators.
  • domain assumption The discretized system matrices A are symmetric positive definite (SPD).
    Stated in Section 2.3 and required for the energy norm analysis, CG, and spectral clustering arguments.
  • standard math Attention weights in NAMG form a Monte Carlo approximation of an integral kernel.
    Theorem 4.4 proof in Appendix C.4 relies on the standard interpretation of softmax attention as a kernel integral, as in [4, 11, 28, 17].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Neural Preconditioning Operator for Efficient PDE Solves." pith.science (2026). https://pith.science/paper/3SGRJHHK

@misc{pith2026250201337,
  author       = {Pith},
  title        = {Pith review of: Neural Preconditioning Operator for Efficient PDE Solves},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3SGRJHHK}},
  note         = {Machine review of arXiv:2502.01337}
}
read the original abstract

We introduce the Neural Preconditioning Operator (NPO), a novel approach designed to accelerate Krylov solvers in solving large, sparse linear systems derived from partial differential equations (PDEs). Unlike classical preconditioners that often require extensive tuning and struggle to generalize across different meshes or parameters, NPO employs neural operators trained via condition and residual losses. This framework seamlessly integrates with existing neural network models, serving effectively as a preconditioner to enhance the performance of Krylov subspace methods. Further, by melding algebraic multigrid principles with a transformer-based architecture, NPO significantly reduces iteration counts and runtime for solving Poisson, Diffusion, and Linear Elasticity problems on both uniform and irregular meshes. Our extensive numerical experiments demonstrate that NPO outperforms traditional methods and contemporary neural approaches across various resolutions, ensuring robust convergence even on grids as large as 4096, far exceeding its initial training limits. These findings underscore the potential of data-driven preconditioning to transform the computational efficiency of high-dimensional PDE applications.

Figures

Figures reproduced from arXiv: 2502.01337 by the authors.

Figure 1
Figure 1. Illustration of Neural Preconditioning Operator Framework. (a) The training phase with multiple loss functions (data, residual, and condition losses) and (b) the solving phase integrated with Krylov subspace methods for efficient PDE solutions. 2.2 Preconditioner Preconditioning improves the convergence of Krylov methods by transforming the system Ax = b into one with more favorable spectral properties: MAx = Mb, (4… view at source ↗
Figure 2
Figure 2. Illustration of Neural Algebraic Multigrid Operator. Theorem 4.1 (Two-Grid Convergence). Let e (k) be the error at iteration k of a two-grid scheme for the SPD system Ax = b. Suppose the coarse correction satisfies the Approximation Property and the smoothing step remains stable. Then there exists a constant ρ < 1 such that ∥e (k+1)∥a ≤ ρ ∥e (k) ∥a, where ∥ · ∥a is the energy norm induced by A. Consequently, the ite… view at source ↗
Figure 3
Figure 3. Relative residual convergence comparison of different solvers for the Poisson equation on a [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Performance comparison of numerical methods across grid resolutions from 128 to 4096. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. RAPNet: Accelerating Algebraic Multigrid with Learned Sparse Corrections

    cs.LG 2026-05 unverdicted novelty 7.0 of 10

    RAPNet uses a GNN with level-wise training to learn sparse robust coarse operators that accelerate algebraic multigrid on large PDE and graph problems.

  2. Neural operator preconditioning from mixed dataset for the Helmholtz equations: Application to transcranial ultrasound

    math.NA 2026-07 conditional novelty 5.0 of 10

    Six mixed training datasets are compared for a U-Net Helmholtz preconditioner; the best mix lets FGMRES solve 512×512 head-CT problems that GMRES and the Stanziola learned optimizer cannot.

Reference graph

Works this paper leans on

29 extracted references · 19 canonical work pages · cited by 2 Pith papers

  1. [1]

    Düben, Tim N

    Jan Ackmann, Peter D. Düben, Tim N. Palmer, and Piotr K. Smolarkiewicz. Machine-learned preconditioners for linear solvers in geophysical fluid flows. CoRR, abs/2010.02866, 2020

    arXiv 2010

  2. [2]

    Briggs, V .E

    W.L. Briggs, V .E. Henson, and S.F. McCormick. A Multigrid Tutorial, second ed. SIAM, California, 2000

    work page 2000

  3. [3]

    Neural-network preconditioners for solving the Dirac equation in lattice gauge theory

    Salvatore Calì, Daniel C. Hackett, Yin Lin, Phiala E. Shanahan, and Brian Xiao. Neural-network preconditioners for solving the dirac equation in lattice gauge theory. CoRR, abs/2208.02728, 2022

    work page Pith review arXiv 2022

  4. [4]

    Choose a transformer: Fourier or galerkin

    Shuhao Cao. Choose a transformer: Fourier or galerkin. In Neural Information Processing Systems, 2021

    work page 2021

  5. [5]

    Finite volume methods

    Robert Eymard, Thierry Gallouët, and Raphaèle Herbin. Finite volume methods. In Solution of Equation in Rn (Part 3), Techniques of Scientific Computing (Part 3), volume 7 of Hand- book of Numerical Analysis , pages 713–1018. Elsevier, 2000. doi: https://doi.org/10.1016/ S1570-8659(00)07005-8

    work page

  6. [6]

    David Gottlieb and Steven A. Orszag. Numerical Analysis of Spectral Methods. Society for Industrial and Applied Mathematics, 1977. doi: 10.1137/1.9781611970425

    doi:10.1137/1.9781611970425 1977

  7. [7]

    Morton E. Gurtin. An Introduction to Continuum Mechanics, Mathematics in Science and Engineering. Academic Press, Cambridge, 1982

    work page 1982

  8. [8]

    Learn- ing neural PDE solvers with convergence guarantees

    Jun-Ting Hsieh, Shengjia Zhao, Stephan Eismann, Lucia Mirabella, and Stefano Ermon. Learn- ing neural PDE solvers with convergence guarantees. In ICLR (Poster). OpenReview.net, 2019

    work page 2019

Show all 29 references
  1. [9]

    Numerical Solution of Partial Differential Equations by the Finite Element Method

    Claes Johnson. Numerical Solution of Partial Differential Equations by the Finite Element Method. Courier Corporation, North Chelmsford, 2012

    2012

  2. [10]

    Deeponet based preconditioning strategies for solving parametric linear systems of equations

    Alena Kopanicáková and George Em Karniadakis. Deeponet based preconditioning strategies for solving parametric linear systems of equations. CoRR, abs/2401.02016, 2024

    2024 arXiv

  3. [11]

    Kovachki, Zongyi Li, Burigede Liu, Kamyar Azizzadenesheli, Kaushik Bhattacharya, Andrew M

    Nikola B. Kovachki, Zongyi Li, Burigede Liu, Kamyar Azizzadenesheli, Kaushik Bhattacharya, Andrew M. Stuart, and Anima Anandkumar. Neural operator: Learning maps between function spaces with applications to pdes. J. Mach. Learn. Res., 24:89:1–89:97, 2023

    2023

  4. [12]

    Multigrid-augmented deep learning preconditioners for the helmholtz equation using compact implicit layers

    Bar Lerer, Ido Ben-Yair, and Eran Treister. Multigrid-augmented deep learning preconditioners for the helmholtz equation using compact implicit layers. SIAM J. Sci. Comput., 46(5):S123– S144, 2024

    2024

  5. [13]

    Randall J. LeVeque. Finite Difference Methods for Ordinary and Partial Differential Equations. Society for Industrial and Applied Mathematics, 2007. doi: 10.1137/1.9780898717839

    2007 doi

  6. [14]

    Machine learning for preconditioning elliptic equations in porous microstructures: A path to error control

    Kangan Li, Sabit Mahmood Khan, and Yashar Mehmani. Machine learning for preconditioning elliptic equations in porous microstructures: A path to error control. Computer Methods in Applied Mechanics and Engineering , 427:117056, 2024. ISSN 0045-7825. doi: https: //doi.org/10.101...

    2024

  7. [15]

    Learning preconditioners for conjugate gradient PDE solvers

    Yichen Li, Peter Yichen Chen, Tao Du, and Wojciech Matusik. Learning preconditioners for conjugate gradient PDE solvers. In ICML, volume 202 of Proceedings of Machine Learning Research, pages 19425–19439. PMLR, 2023

    2023

  8. [16]

    M2NO: multiresolution operator learning with multiwavelet-based algebraic multigrid method

    Zhihao Li, Zhilu Lai, Xiaobo Zhang, and Wei Wang. M2NO: multiresolution operator learning with multiwavelet-based algebraic multigrid method. CoRR, abs/2406.04822, 2024

    2024

  9. [17]

    Harnessing scale and physics: A multi-graph neural operator framework for pdes on arbitrary geometries.CoRR, abs/2411.15178, 2024

    Zhihao Li, Haoze Song, Di Xiao, Zhilu Lai, and Wei Wang. Harnessing scale and physics: A multi-graph neural operator framework for pdes on arbitrary geometries.CoRR, abs/2411.15178, 2024

    2024 arXiv

  10. [18]

    Stuart, and Anima Anandkumar

    Zongyi Li, Nikola Borislavov Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhat- tacharya, Andrew M. Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differential equations. In ICLR. OpenReview.net, 2021

    2021

  11. [19]

    Learning nonlinear operators via deeponet based on the universal approximation theorem of operators

    Lu Lu, Pengzhan Jin, Guofei Pang, Zhongqiang Zhang, and George Em Karniadakis. Learning nonlinear operators via deeponet based on the universal approximation theorem of operators. Nat. Mach. Intell., 3(3):218–229, 2021

    2021

  12. [20]

    Brenner, and Shmuel M

    Ryan McKeown, Rodolfo Ostilla-Mónico, Alain Pumir, Michael P. Brenner, and Shmuel M. Rubinstein. Turbulence generation through an iterative cascade of the elliptical instability. Science Advances, 6(9):eaaz2717, 2020. doi: 10.1126/sciadv.aaz2717

    2020 doi

  13. [21]

    Jorge Nocedal and Stephen J. Wright. Numerical Optimization. Springer Series in Operations Research and Financial Engineering. Springer New York, NY , 2 edition, 2006. ISBN 978-0- 387-30303-1. doi: 10.1007/978-0-387-40065-5

    2006 doi

  14. [22]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In MICCAI (3), volume 9351 of Lecture Notes in Computer Science, pages 234–241. Springer, 2015

    2015

  15. [23]

    Muravleva, Yuri M

    Alexander Rudikov, Vladimir Fanaskov, Ekaterina A. Muravleva, Yuri M. Laevsky, and Ivan V . Oseledets. Neural operators meet conjugate gradients: The FCG-NO method for efficient PDE solving. In ICML. OpenReview.net, 2024

    2024

  16. [24]

    Rumelhart, Geoffrey E

    David E. Rumelhart, Geoffrey E. Hinton, and Ronald J. Williams. Learning representations by back-propagating errors. Nature, 323:533–536, 1986

    1986

  17. [25]

    M. F. Shlesinger, B. J. West, and J. Klafter. Lévy dynamics of enhanced diffusion: Application to turbulence. Phys. Rev. Lett., 58:1100–1103, Mar 1987. doi: 10.1103/PhysRevLett.58.1100

    1987 doi

  18. [26]

    Computational Science and Engineering

    Gilbert Strang. Computational Science and Engineering. Wellesley-Cambridge Press, Philadel- phia, PA, 2007. doi: 10.1137/1.9780961408817

    2007 doi

  19. [27]

    Partial Differential Equations: Methods and Applications

    Abdul-Majid Wazwaz. Partial Differential Equations: Methods and Applications. Balkema Publishers, Leiden, 2002

    2002

  20. [28]

    Transolver: A fast transformer solver for pdes on general geometries

    Haixu Wu, Huakun Luo, Haowen Wang, Jianmin Wang, and Mingsheng Long. Transolver: A fast transformer solver for pdes on general geometries. ArXiv, abs/2402.02366, 2024

    2024 arXiv

  21. [29]

    best fit

    Enrui Zhang, Adar Kahana, Eli Turkel, Rishikesh Ranade, Jay Pathak, and George Em Kar- niadakis. A hybrid iterative numerical transferable solver (HINTS) for pdes based on deep operator network and relaxation methods. CoRR, abs/2208.13273, 2022. A Related Work A.1 Numerical Pr...

    2022 arXiv

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.