pith. sign in

arxiv: 2606.18305 · v1 · pith:NKGTH25Anew · submitted 2026-06-16 · 🧮 math.NA · cs.LG· cs.NA

Starter-Iterator Neural Operator: A Unified Architecture for High-Fidelity Forward and Inverse PDE Problems

Kuilin Qin , Lianfang Wang , Xu Sun , Jiwei Jia , Yu Wang , Yong Wang , Yuping Duan This is my paper

Pith reviewed 2026-06-27 00:12 UTC · model grok-4.3

classification 🧮 math.NA cs.LGcs.NA
keywords neural operatorsoperator learningPDE surrogate modelingforward and inverse problemsspectral methodsiterative methodsdynamical systems
0
0 comments X

The pith

The Starter-Iterator Neural Operator combines frequency-domain initialization with time-domain iteration to solve forward and inverse PDE problems more accurately than single-domain methods.

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

The paper introduces the Starter-Iterator Neural Operator to map infinite-dimensional function spaces for high-dimensional PDEs, offering an efficient surrogate that balances computational cost and accuracy better than traditional solvers. It reinterprets the initialization and iteration steps of classical iterative methods as neural network components to enable spectral-spatiotemporal collaborative modeling. A frequency-domain module captures globally stable low-frequency features while a time-domain module refines local residuals, addressing precision limits on complex boundaries and long-term evolution. Experiments on Navier-Stokes equations, acoustic wave equations, super-resolution imaging, and weather forecasting report gains in numerical accuracy, generalization, and robustness for both forward and inverse tasks.

Core claim

SINO reinterprets initialization strategies and iterative formats of traditional iterative methods through neural networks to create an efficient spectral-spatiotemporal collaborative modeling approach, where the frequency-domain initialization module captures globally stable low-frequency features and the time-domain learning module focuses on optimizing local solution residuals.

What carries the argument

Frequency-domain initialization module paired with time-domain learning module inside the Starter-Iterator Neural Operator for spectral-spatiotemporal collaborative modeling of PDE operators.

If this is right

  • The architecture supplies a unified framework that meets the stringent accuracy needs of both forward simulation and inverse inference within a single model.
  • It delivers a better trade-off between computational complexity and approximation accuracy for many-query tasks such as real-time prediction and parameter sweeps.
  • Practical applications including super-resolution imaging and weather forecasting gain measurable improvements in robustness and generalization.
  • The dual-module design directly mitigates precision bottlenecks that arise when complex boundaries or extended time horizons are present.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same frequency-plus-time decomposition could be tested on other classes of operator-learning problems that currently suffer from domain-specific drift.
  • If the modules prove additive, hybrid initialization schemes might be explored for additional dynamical systems not covered in the reported experiments.
  • The unified forward-inverse capability suggests potential use in closed-loop control or data-assimilation pipelines that alternate between the two problem types.

Load-bearing premise

The frequency-domain initialization and time-domain learning modules effectively overcome the limitations of single-domain modeling for complex boundaries and long-term evolution.

What would settle it

A controlled test on long-term evolution of the Navier-Stokes equations in which SINO shows no accuracy or generalization improvement over existing operator learning baselines would falsify the central performance claim.

Figures

Figures reproduced from arXiv: 2606.18305 by Jiwei Jia, Kuilin Qin, Lianfang Wang, Xu Sun, Yong Wang, Yuping Duan, Yu Wang.

Figure 1
Figure 1. Figure 1: Illustration of the proposed starter-iterator neural operator. (a) General framework of operator learning, where the input in space is lifted into latent source space X , processed by the neural operator to get the target latent space Y, and projected to true solution space as the output. (b) Framework of our proposed Starter-Iterator Neural Operator (SINO), which incorporates both Starter and Iterator mod… view at source ↗
Figure 2
Figure 2. Figure 2: Relative L2 error comparison for zero-shot resolution generalization on the 1D Burgers equation. We trained the FNO and SINO with different iterator steps I only on low-resolution data and directly evaluated on higher-resolution test sets without any retraining or fine-tuning. more pronounced as the number of iterator steps increases. In particular, SINO with iterator steps I = 32 achieves the lowest error… view at source ↗
Figure 3
Figure 3. Figure 3: Hyperparameter analysis on the Darcy-flow benchmark. [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Effect of the starter module and multiscale depth on model convergence for operator learning of the Darcy PDE. The left panel compares [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Spectral phenomenon of SINO. (a) Training loss curves for both low- and high-frequency components under different initialization frequency combinations. (b) Visualization of the four sets of harmonic functions corresponding to the frequency bands selected in the experiments. (c) Qualitative fitting results at epochs 20, 50, and 80, using three out of the four frequency groups from (b) for initialization. T… view at source ↗
Figure 6
Figure 6. Figure 6: (a) Heatmap of training losses with respect to different input and output sequence lengths [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Results on Navier-Stokes equation and Wave equation. The first and second columns show predicted solutions (first column) and corresponding error maps (second column) for the 2D Navier–Stokes equation with viscosity ν = 1e − 3, generated by different operator learning methods (one per row). The third and fourth columns display analogous results under reduced viscosity ν = 1e − 4. The fifth and sixth column… view at source ↗
Figure 8
Figure 8. Figure 8: Visualization of residual maps between predicted and ground-truth solutions of the shallow-water equation. Rows 1–4 correspond the [PITH_FULL_IMAGE:figures/full_fig_p020_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Quantitative and qualitative evaluation of ×2 super-resolution performance across multiple datasets. (a) Representative visual comparisons of ×2 super-resolution results from different methods on datasets with increasing structural complexity from top to bottom. For each dataset, columns show the widefield (WF) input, predictions from DFCAN, RCAN, and our method, followed by the ground truth structured ill… view at source ↗
Figure 10
Figure 10. Figure 10: Visualization of training dynamics on four datasets. The x-axis represents the number of epoch and the y-axis the error in log scale. [PITH_FULL_IMAGE:figures/full_fig_p023_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: The comparison of Weather prediction results. (a) Monthly prediction loss over the year 2018 comparing the proposed surrogate model (red) with the baseline model (black); (b) prediction loss curves for June; (c) visualization of selected time steps in June, where the first row shows the ground truth, the second row the predictions from SINO, the third row the predictions from the baseline model, and the f… view at source ↗
read the original abstract

Operator learning is an emerging interdisciplinary field that integrates machine learning with scientific computing. By mapping infinite-dimensional function spaces, this approach provides an efficient surrogate modeling framework for high-dimensional partial differential equations (PDEs). Compared to traditional numerical solvers, it achieves a superior trade-off between computational complexity and approximation accuracy, demonstrating significant advantages in many-query tasks such as real-time prediction and parameter sweeps. Given the stringent accuracy requirements of both forward simulation and inverse inference, as well as the precision bottlenecks of existing operator learning methods in handling complex boundaries or long-term evolution, we propose the Starter-Iterator Neural Operator (SINO). Our framework reinterprets the initialization strategies and iterative formats of traditional iterative methods through neural networks, establishing an efficient approach for spectral-spatiotemporal collaborative modeling. Specifically, the frequency-domain initialization module captures globally stable low-frequency features, while the time-domain learning module focuses on optimizing local solution residuals, thereby effectively overcoming the inherent limitations of conventional single-domain modeling approaches. Extensive experiments on typical dynamical systems such as the Navier-Stokes equations and acoustic wave equations, as well as practical applications including super-resolution imaging and weather forecasting, demonstrate that SINO achieves outstanding performance in numerical accuracy, generalization capability, and robustness.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The manuscript proposes the Starter-Iterator Neural Operator (SINO), a unified neural architecture for high-fidelity forward and inverse PDE problems. It reinterprets initialization and iteration from traditional solvers via neural networks, using a frequency-domain initialization module to capture globally stable low-frequency features and a time-domain learning module to optimize local residuals for spectral-spatiotemporal collaborative modeling. The central claim is that this overcomes limitations of single-domain approaches for complex boundaries and long-term evolution, with extensive experiments on Navier-Stokes equations, acoustic wave equations, super-resolution imaging, and weather forecasting demonstrating outstanding numerical accuracy, generalization, and robustness.

Significance. If the performance claims hold with rigorous validation, SINO could advance operator learning by offering an efficient surrogate framework that improves the accuracy-complexity trade-off for many-query PDE tasks. The spectral-spatiotemporal split addresses a recognized gap in existing methods and, if substantiated, would strengthen the case for hybrid domain modeling in scientific machine learning.

minor comments (3)
  1. The abstract asserts 'outstanding performance' and 'extensive experiments' without any quantitative error metrics, baseline comparisons, or dataset details; adding a concise summary of key results (e.g., relative L2 errors versus FNO or DeepONet) would strengthen the summary paragraph.
  2. Notation for the frequency-domain and time-domain modules is introduced at a high level; a dedicated subsection or diagram clarifying the exact network architectures, activation functions, and how the starter and iterator components are combined would improve reproducibility.
  3. The manuscript should explicitly state the training loss, optimizer settings, and hyperparameter ranges used across all experiments to allow direct comparison with prior operator-learning work.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of our manuscript and the recommendation for minor revision. The recognition of SINO's potential to advance hybrid domain modeling in scientific machine learning is appreciated.

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper presents SINO as a proposed neural architecture that reinterprets traditional iterative methods via neural networks for spectral-spatiotemporal modeling of PDEs. No equations, uniqueness theorems, or fitted parameters are shown to reduce by construction to the inputs or to self-citations; performance claims rest on experimental validation across Navier-Stokes, wave equations, and applications rather than any self-definitional or load-bearing derivation step. The architecture description and results are self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The abstract does not specify any free parameters, axioms, or invented entities; a full text review would be needed to identify them.

pith-pipeline@v0.9.1-grok · 5765 in / 942 out tokens · 40625 ms · 2026-06-27T00:12:49.474407+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

74 extracted references · 3 canonical work pages · 1 internal anchor

  1. [1]

    J. Li, Y . Xue, Y . Li, H. Jia, Z. Zhou, L. Yang, S. Ren, J. Chen, Y . He, K. Xue, et al., Fully analog iteration for solving matrix equations with in-memory computing, Science Advances 11 (7) (2025) eadr6391

    2025

  2. [2]

    Y . Tang, R. Chen, M. Lou, J. Fan, C. Yu, A. Nonaka, Z. Yao, W. Gao, Optical neural engine for solving scientific partial differential equations, Nature Communications 16 (1) (2025) 4603

    2025

  3. [3]

    K. Ding, J. Yu, J. Huang, Y . Yang, Q. Zhang, H. Chen, Scitoolagent: a knowledge-graph-driven scientific agent for multitool integration, Nature Computational Science (2025) 1–11

    2025

  4. [4]

    W. He, J. Li, X. Kong, L. Deng, Multi-level physics informed deep learning for solving partial differential equations in computational structural mechanics, Communications Engineering 3 (1) (2024) 151

    2024

  5. [5]

    K. L. Tsakmakidis, T. P. Stefa ´nski, Discovery of the exact 3d one-way wave equation, Nature Communications 16 (1) (2025) 5719

    2025

  6. [6]

    Allen, S

    A. Allen, S. Markou, W. Tebbutt, J. Requeima, W. P. Bruinsma, T. R. Andersson, M. Herzog, N. D. Lane, M. Chantry, J. S. Hosking, et al., End-to-end data-driven weather prediction, Nature 641 (8065) (2025) 1172– 1179

    2025

  7. [7]

    Bauer, P

    P. Bauer, P. Dueben, M. Chantry, F. Doblas-Reyes, T. Hoefler, A. McGovern, B. Stevens, Deep learning and a changing economy in weather and climate prediction, Nature Reviews Earth & Environment 4 (8) (2023) 507– 509

    2023

  8. [8]

    Aarts, K

    G. Aarts, K. Fukushima, T. Hatsuda, A. Ipp, S. Shi, L. Wang, K. Zhou, Physics-driven learning for inverse problems in quantum chromodynamics, Nature Reviews Physics 7 (3) (2025) 154–163

    2025

  9. [9]

    Tarantola, Inverse problem theory and methods for model parameter estimation, SIAM (2005)

    A. Tarantola, Inverse problem theory and methods for model parameter estimation, SIAM (2005)

    2005

  10. [10]

    C. Lyu, B. Romanowicz, L. Zhao, Y . Masson, Efficient hybrid numerical modeling of the seismic wavefield in the presence of solid-fluid boundaries, Nature Communications 16 (1) (2025) 1722. 31

    2025

  11. [11]

    H. Zhao, Z. Liu, J. Tang, B. Gao, Q. Qin, J. Li, Y . Zhou, P. Yao, Y . Xi, Y . Lin, et al., Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis, Nature Communications 14 (1) (2023) 2276

    2023

  12. [12]

    Carleo, I

    G. Carleo, I. Cirac, K. Cranmer, L. Daudet, M. Schuld, N. Tishby, L. V ogt-Maranto, L. Zdeborová, Machine learning and the physical sciences, Reviews of Modern Physics 91 (4) (2019) 045002

    2019

  13. [13]

    S. L. Brunton, J. N. Kutz, Promising directions of machine learning for partial differential equations, Nature Computational Science 4 (7) (2024) 483–494

    2024

  14. [14]

    Kontolati, S

    K. Kontolati, S. Goswami, G. Em Karniadakis, M. D. Shields, Learning nonlinear operators in latent spaces for real-time predictions of complex dynamics in physical systems, Nature Communications 15 (1) (2024) 5101

    2024

  15. [15]

    Zappala, A

    E. Zappala, A. H. d. O. Fonseca, J. O. Caro, A. H. Moberly, M. J. Higley, J. Cardin, D. v. Dijk, Learning integral operators via neural integral equations, Nature Machine Intelligence 6 (9) (2024) 1046–1062

    2024

  16. [16]

    Georgiou, C

    K. Georgiou, C. Siettos, A. N. Yannacopoulos, Fredholm neural networks, SIAM Journal on Scientific Comput- ing 47 (4) (2025) C1006–C1031

    2025

  17. [17]

    Azizzadenesheli, N

    K. Azizzadenesheli, N. Kovachki, Z. Li, M. Liu-Schiaffini, J. Kossaifi, A. Anandkumar, Neural operators for accelerating scientific simulations and design, Nature Reviews Physics 6 (5) (2024) 320–328

    2024

  18. [18]

    B. Wu, C. Liu, B. Eckart, J. Kautz, Neural interferometry: Image reconstruction from astronomical interferom- eters using transformer-conditioned neural fields, in: Proceedings of the AAAI Conference on Artificial Intelli- gence, V ol. 36, 2022, pp. 2685–2693

    2022

  19. [19]

    J. N. Reddy, An introduction to the finite element method, New York 27 (14) (1993)

    1993

  20. [20]

    G. D. Smith, Numerical solution of partial differential equations: finite difference methods, Oxford University Press (1985)

    1985

  21. [21]

    Eymard, T

    R. Eymard, T. Gallouët, R. Herbin, Finite volume methods, Handbook of numerical analysis 7 (2000) 713–1018

    2000

  22. [22]

    J. Shen, T. Tang, L.-L. Wang, Spectral methods: algorithms, analysis and applications, Springer Science & Business Media (2011)

    2011

  23. [23]

    Zhang, A

    E. Zhang, A. Kahana, A. Kopani ˇcáková, E. Turkel, R. Ranade, J. Pathak, G. E. Karniadakis, Blending neural operators and relaxation methods in pde numerical solvers, Nature Machine Intelligence 6 (11) (2024) 1303– 1313

    2024

  24. [24]

    A. Jiao, H. He, R. Ranade, J. Pathak, L. Lu, One-shot learning for solution operators of partial differential equations, Nature Communications 16 (1) (2025) 8386

    2025

  25. [25]

    J. M. Ortega, M. L. Rockoff, Nonlinear difference equations and gauss-seidel type iterative methods, SIAM Journal on Numerical Analysis 3 (3) (1966) 497–513

    1966

  26. [26]

    Saad, Krylov subspace methods for solving large unsymmetric linear systems, Mathematics of computation 37 (155) (1981) 105–126

    Y . Saad, Krylov subspace methods for solving large unsymmetric linear systems, Mathematics of computation 37 (155) (1981) 105–126

    1981

  27. [27]

    M. R. Hestenes, E. Stiefel, et al., Methods of conjugate gradients for solving linear systems, Journal of research of the National Bureau of Standards 49 (6) (1952) 409–436

    1952

  28. [28]

    Y . Saad, M. H. Schultz, Gmres: A generalized minimal residual algorithm for solving nonsymmetric linear systems, SIAM Journal on scientific and statistical computing 7 (3) (1986) 856–869

    1986

  29. [29]

    A. H. Sherman, On newton-iterative methods for the solution of systems of nonlinear equations, SIAM Journal on Numerical Analysis 15 (4) (1978) 755–771

    1978

  30. [30]

    Molinaro, Y

    R. Molinaro, Y . Yang, B. Engquist, S. Mishra, Neural inverse operators for solving pde inverse problems, in: Proceedings of the 40th International Conference on Machine Learning, ICML’23, JMLR.org, 2023

    2023

  31. [31]

    Z. Li, H. Zheng, N. Kovachki, D. Jin, H. Chen, B. Liu, K. Azizzadenesheli, A. Anandkumar, Physics-informed neural operator for learning partial differential equations, ACM/IMS Journal of Data Science 1 (3) (2024) 1–27. 32

    2024

  32. [32]

    X. Liu, H. Tang, Difffno: Diffusion fourier neural operator, in: Proceedings of the Computer Vision and Pattern Recognition Conference, 2025, pp. 150–160

    2025

  33. [33]

    Kovachki, Z

    N. Kovachki, Z. Li, B. Liu, K. Azizzadenesheli, K. Bhattacharya, A. Stuart, A. Anandkumar, Neural operator: Learning maps between function spaces with applications to pdes, Journal of Machine Learning Research 24 (89) (2023) 1–97

    2023

  34. [34]

    J. S. Hesthaven, S. Ubbiali, Non-intrusive reduced order modeling of nonlinear problems using neural networks, Journal of Computational Physics 363 (2018) 55–78

    2018

  35. [35]

    Bhattacharya, B

    K. Bhattacharya, B. Hosseini, N. B. Kovachki, A. M. Stuart, Model reduction and neural networks for parametric pdes, The SMAI journal of computational mathematics 7 (2021) 121–157

    2021

  36. [36]

    O’Leary-Roseberry, U

    T. O’Leary-Roseberry, U. Villa, P. Chen, O. Ghattas, Derivative-informed projected neural networks for high- dimensional parametric maps governed by pdes, Computer Methods in Applied Mechanics and Engineering 388 (2022) 114199

    2022

  37. [37]

    L. Lu, P. Jin, G. Pang, Z. Zhang, G. E. Karniadakis, Learning nonlinear operators via deeponet based on the universal approximation theorem of operators, Nature machine intelligence 3 (3) (2021) 218–229

    2021

  38. [38]

    J. He, S. Koric, D. Abueidda, A. Najafi, I. Jasiuk, Geom-deeponet: A point-cloud-based deep operator network for field predictions on 3d parameterized geometries, Computer Methods in Applied Mechanics and Engineering 429 (2024) 117130

    2024

  39. [39]

    W. Xu, Y . Lu, L. Wang, Transfer learning enhanced deeponet for long-time prediction of evolution equations, in: Proceedings of the AAAI Conference on Artificial Intelligence, V ol. 37, 2023, pp. 10629–10636

    2023

  40. [40]

    L. Lu, X. Meng, S. Cai, Z. Mao, S. Goswami, Z. Zhang, G. E. Karniadakis, A comprehensive and fair comparison of two neural operators (with practical extensions) based on fair data, Computer Methods in Applied Mechanics and Engineering 393 (2022) 114778

    2022

  41. [41]

    Z. Li, N. B. Kovachki, K. Azizzadenesheli, B. liu, K. Bhattacharya, A. Stuart, A. Anandkumar, Fourier neural operator for parametric partial differential equations, in: International Conference on Learning Representations, 2021

    2021

  42. [42]

    V . S. Fanaskov, I. V . Oseledets, Spectral neural operators, in: Doklady Mathematics, V ol. 108, Springer, 2023, pp. S226–S232

    2023

  43. [43]

    Tripura, S

    T. Tripura, S. Chakraborty, Wavelet neural operator for solving parametric partial differential equations in com- putational mechanics problems, Computer Methods in Applied Mechanics and Engineering 404 (2023) 115783

    2023

  44. [44]

    Q. Cao, S. Goswami, G. E. Karniadakis, Laplace neural operator for solving differential equations, Nature Ma- chine Intelligence 6 (6) (2024) 631–640

    2024

  45. [45]

    K. Li, W. Ye, D-fno: A decomposed fourier neural operator for large-scale parametric partial differential equa- tions, Computer Methods in Applied Mechanics and Engineering 436 (2025) 117732

    2025

  46. [46]

    Lehmann, F

    F. Lehmann, F. Gatti, M. Bertin, D. Clouteau, 3d elastic wave propagation with a factorized fourier neural operator (f-fno), Computer Methods in Applied Mechanics and Engineering 420 (2024) 116718

    2024

  47. [47]

    M. A. Rahman, Z. E. Ross, K. Azizzadenesheli, U-NO: U-shaped neural operators, Transactions on Machine Learning Research (2023)

    2023

  48. [48]

    J. He, X. Liu, J. Xu, Mgno: Efficient parameterization of linear operators via multigrid, in: The Twelfth Interna- tional Conference on Learning Representations, 2024

    2024

  49. [49]

    B. M. DeBlois, Linearizing convection terms in the navier-stokes equations, Computer methods in applied me- chanics and engineering 143 (3-4) (1997) 289–297

    1997

  50. [50]

    J. M. Borwein, G. Li, M. K. Tam, Convergence rate analysis for averaged fixed point iterations in common fixed point problems, SIAM Journal on Optimization 27 (1) (2017) 1–33. 33

    2017

  51. [51]

    J. Ho, A. Jain, P. Abbeel, Denoising diffusion probabilistic models, Advances in neural information processing systems 33 (2020) 6840–6851

    2020

  52. [52]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, I. Polosukhin, Attention is all you need, Advances in neural information processing systems 30 (2017)

    2017

  53. [53]

    Stevens, L

    E. Stevens, L. Antiga, T. Viehmann, Deep Learning with PyTorch: Build, train, and tune neural networks using Python tools, Manning, 2020

    2020

  54. [54]

    Xu, Iterative methods by space decomposition and subspace correction, SIAM review 34 (4) (1992) 581–613

    J. Xu, Iterative methods by space decomposition and subspace correction, SIAM review 34 (4) (1992) 581–613

    1992

  55. [55]

    Z. Zeng, Y . Zheng, H. Hu, Z. Dong, Y . Zheng, X. Liu, J. Wang, Z. Shi, L. Zhang, Y . Li, et al., Openbreastus: Benchmarking neural operators for wave imaging using breast ultrasound computed tomography, arXiv preprint arXiv:2507.15035 (2025)

    work page arXiv 2025

  56. [56]

    Takamoto, T

    M. Takamoto, T. Praditia, R. Leiteritz, D. MacKinlay, F. Alesiani, D. Pflüger, M. Niepert, Pdebench: An exten- sive benchmark for scientific machine learning, Advances in Neural Information Processing Systems 35 (2022) 1596–1611

    2022

  57. [57]

    M. J. Rust, M. Bates, X. Zhuang, Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (storm), Nature methods 3 (10) (2006) 793–796

    2006

  58. [58]

    Betzig, G

    E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S. Olenych, J. S. Bonifacino, M. W. Davidson, J. Lippincott-Schwartz, H. F. Hess, Imaging intracellular fluorescent proteins at nanometer resolution, science 313 (5793) (2006) 1642–1645

    2006

  59. [59]

    D. Li, L. Shao, B.-C. Chen, X. Zhang, M. Zhang, B. Moses, D. E. Milkie, J. R. Beach, J. A. Hammer III, M. Pasham, et al., Extended-resolution structured illumination imaging of endocytic and cytoskeletal dynamics, Science 349 (6251) (2015) aab3500

    2015

  60. [60]

    C. Qiao, D. Li, Y . Guo, C. Liu, T. Jiang, Q. Dai, D. Li, Evaluation and development of deep neural networks for image super-resolution in optical microscopy, Nature methods 18 (2) (2021) 194–202

    2021

  61. [61]

    Zhang, K

    Y . Zhang, K. Li, K. Li, L. Wang, B. Zhong, Y . Fu, Image super-resolution using very deep residual channel attention networks, in: Proceedings of the European conference on computer vision (ECCV), 2018, pp. 286– 301

    2018

  62. [62]

    L. Chen, X. Zhong, F. Zhang, Y . Cheng, Y . Xu, Y . Qi, H. Li, Fuxi: a cascade machine learning forecasting system for 15-day global weather forecast, npj climate and atmospheric science 6 (1) (2023) 190

    2023

  63. [63]

    K. Bi, L. Xie, H. Zhang, X. Chen, X. Gu, Q. Tian, Accurate medium-range global weather forecasting with 3d neural networks, Nature 619 (7970) (2023) 533–538

    2023

  64. [64]

    Kurth, S

    T. Kurth, S. Subramanian, P. Harrington, J. Pathak, M. Mardani, D. Hall, A. Miele, K. Kashinath, A. Anandku- mar, Fourcastnet: Accelerating global high-resolution weather forecasting using adaptive fourier neural opera- tors, in: Proceedings of the platform for advanced scientific computing conference, 2023, pp. 1–11

    2023

  65. [65]

    K. Lin, X. Li, Y . Ye, S. Feng, B. Zhang, G. Xu, Z. Wang, Spherical neural operator network for global weather prediction, IEEE Transactions on Circuits and Systems for Video Technology 34 (6) (2023) 4899–4913

    2023

  66. [66]

    S. Rasp, P. D. Dueben, S. Scher, J. A. Weyn, S. Mouatadid, N. Thuerey, Weatherbench: a benchmark data set for data-driven weather forecasting, Journal of Advances in Modeling Earth Systems 12 (11) (2020) e2020MS002203

    2020

  67. [67]

    Gilton, G

    D. Gilton, G. Ongie, R. Willett, Deep equilibrium architectures for inverse problems in imaging, IEEE Transac- tions on Computational Imaging 7 (2021) 1123–1133

    2021

  68. [68]

    Y . Yang, Q. Gao, Y . Duan, Low-resolution prior equilibrium network for ct reconstruction, Inverse Problems 40 (8) (2024) 085010. 34

    2024

  69. [69]

    T. Chen, H. Chen, Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems, IEEE transactions on neural networks 6 (4) (1995) 911–917

    1995

  70. [70]

    Kovachki, S

    N. Kovachki, S. Lanthaler, S. Mishra, On universal approximation and error bounds for fourier neural operators, Journal of Machine Learning Research 22 (290) (2021) 1–76

    2021

  71. [71]

    Monga, Y

    V . Monga, Y . Li, Y . C. Eldar, Algorithm unrolling: Interpretable, efficient deep learning for signal and image processing, IEEE Signal Processing Magazine 38 (2) (2021) 18–44

    2021

  72. [72]

    Z.-Q. J. Xu, Y . Zhang, Z. Zhou, An overview of condensation phenomenon in deep learning, arXiv preprint arXiv:2504.09484 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  73. [73]

    Z. Zeng, Y . Zheng, Y . Zheng, Y . Li, Z. Shi, H. Sun, Neural born series operator for biomedical ultrasound computed tomography, arXiv preprint arXiv:2312.15575 (2023)

    work page arXiv 2023

  74. [74]

    D. I. Ketcheson, K. Mandli, A. J. Ahmadia, A. Alghamdi, M. Q. De Luna, M. Parsani, M. G. Knepley, M. Em- mett, Pyclaw: Accessible, extensible, scalable tools for wave propagation problems, SIAM Journal on Scientific Computing 34 (4) (2012) C210–C231. 35 Figure D.12: Summary of the time-evolution PDE benchmarks considered in this work, including the govern...

    2012