pith. sign in

arxiv: 2607.00178 · v1 · pith:UQ4REIF4new · submitted 2026-06-30 · ⚛️ physics.comp-ph

A Scoping Review of Physics Informed Machine Learning for Wave Propagation Modeling in Seismology

\'Oscar Rinc\'on-Carde\~no , Gregorio P\'erez-Bernal , Silvana Montoya-Noguera , Nicol\'as Guar\'in-Zapata This is my paper

Pith reviewed 2026-07-02 00:46 UTC · model grok-4.3

classification ⚛️ physics.comp-ph
keywords physics-informed machine learningseismologywave propagationscoping reviewforward modelinginverse problemsPINN
0
0 comments X

The pith

Physics-informed machine learning has been applied to seismic wave propagation in both forward modeling and inversion, often matching numerical method accuracy at lower computational cost.

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

This scoping review maps applications of physics-informed machine learning to seismic wave propagation governed by partial differential equations. Selected studies are sorted by problem type, forward or inverse, and by the machine learning strategy employed. The survey finds that these methods have achieved accuracy levels comparable to conventional numerical techniques while incurring lower computational expense. Three mechanisms for embedding physical knowledge are distinguished: observational bias, inductive bias, and learning bias. The review concludes that such approaches remain complementary to standard numerical methods, especially for inverse problems and surrogate modeling, while noting persistent gaps in benchmarking consistency and scaling to three-dimensional or experimentally validated settings.

Core claim

Physics-informed machine learning has been applied to both forward modeling and inversion in seismology, often reaching accuracy comparable to standard numerical methods at lower computational cost. Application of three mechanisms for incorporating physical knowledge were identified: observational bias, inductive bias, and learning bias. Replication of the original PINN framework produced results consistent with and in most cases more accurate than those originally reported.

What carries the argument

Classification of studies by problem type (forward or inverse) and machine learning strategy, together with the three mechanisms for incorporating physical knowledge into the training process.

Load-bearing premise

The chosen databases and classification scheme captured a representative sample of relevant studies without major omissions or selection bias.

What would settle it

A large set of studies on physics-informed machine learning for seismic waves that were not retrieved by the search, or direct comparisons showing that the replicated methods fail to match reported accuracy.

read the original abstract

\emph{Background:} Standard numerical methods accurately simulate seismic waves but are computationally expensive, particularly for inverse problems. Machine learning approaches have been proposed as alternatives that can reduce computational cost while maintaining acceptable physical accuracy. \emph{Objective:} To map how physics-informed machine learning methods have been applied to seismic wave propagation modeling based on partial differential equations. \emph{Methods:} A scoping review was conducted using the OpenAlex and Scopus databases. Selected studies were classified by problem type (forward or inverse) and machine learning strategy to identify research trends, methodological patterns, and gaps in the literature. \emph{Results:} Physics-informed machine learning has been applied to both forward modeling and inversion in seismology, often reaching accuracy comparable to standard numerical methods at lower computational cost. Application of three mechanisms for incorporating physical knowledge were identified: observational bias, inductive bias, and learning bias. To evaluate methodological reproducibility of a representative method, the original PINN framework was replicated in PyTorch, obtaining results consistent with and in most cases more accurate than those originally reported. From the reviewed literature, limitations remain in benchmarking consistency, training cost, and scalability to three-dimensional and experimentally validated problems. \emph{Conclusions:} Standard numerical methods remain the basis of seismological workflows, while physics-informed machine learning offers complementary approaches that are useful for inverse problems and surrogate modeling. Future work should focus on consistent benchmarking, hybrid formulations, and validation under realistic geophysical conditions.

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

2 major / 2 minor

Summary. The manuscript is a scoping review that maps applications of physics-informed machine learning to seismic wave propagation modeling. It reports a literature search in OpenAlex and Scopus, classifies studies by problem type (forward/inverse) and ML strategy (observational, inductive, and learning bias), concludes that these methods often achieve accuracy comparable to numerical methods at lower cost, presents a PyTorch replication of the original PINN framework yielding consistent or improved results, and notes remaining limitations in benchmarking consistency, training cost, and scalability to 3D/experimental cases. The conclusion positions PIML approaches as complementary to standard numerical methods for inverse problems and surrogate modeling.

Significance. If the sampled literature is representative, the review would usefully synthesize trends and gaps in PIML for seismology. The explicit replication of the PINN framework constitutes a concrete strength by demonstrating reproducibility for at least one case. The aggregate performance claim ('often' comparable accuracy at lower cost) would be informative for the field if supported by transparent search and selection procedures.

major comments (2)
  1. [Methods] Methods section (and abstract): the scoping review description provides no search strings, inclusion/exclusion criteria, screening counts, or PRISMA flow diagram. Without these, the representativeness of the OpenAlex/Scopus sample cannot be evaluated, directly undermining the load-bearing claim that PIML methods 'often' reach accuracy comparable to numerical methods at lower computational cost.
  2. [Results] Results section: the classification of studies into observational bias, inductive bias, and learning bias is stated without explicit operational definitions, decision rules, or a table mapping individual papers to categories. This makes it impossible to assess whether the taxonomy is applied consistently or exhaustively across the reviewed set.
minor comments (2)
  1. [Results] The replication is described only at a high level; adding quantitative metrics (e.g., L2 error values, training epochs) comparing the PyTorch implementation to the original would strengthen the reproducibility claim.
  2. Ensure every cited study is accompanied by a stable identifier (DOI or arXiv ID) so readers can locate the primary sources used for the trends and performance statements.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our scoping review. We agree that the Methods and Results sections require additional detail to ensure transparency and reproducibility. We address each major comment below and will incorporate revisions in the next version of the manuscript.

read point-by-point responses

  1. Referee: [Methods] Methods section (and abstract): the scoping review description provides no search strings, inclusion/exclusion criteria, screening counts, or PRISMA flow diagram. Without these, the representativeness of the OpenAlex/Scopus sample cannot be evaluated, directly undermining the load-bearing claim that PIML methods 'often' reach accuracy comparable to numerical methods at lower computational cost.

    Authors: We acknowledge the omission of these procedural details. The literature search was performed using specific queries in OpenAlex and Scopus, followed by screening according to defined inclusion/exclusion criteria, but these elements were not reported in the submitted manuscript. We will expand the Methods section to include the exact search strings, inclusion/exclusion criteria, screening counts, and a PRISMA flow diagram. This addition will allow readers to evaluate sample representativeness and will provide transparent support for the aggregate performance observations. revision: yes

  2. Referee: [Results] Results section: the classification of studies into observational bias, inductive bias, and learning bias is stated without explicit operational definitions, decision rules, or a table mapping individual papers to categories. This makes it impossible to assess whether the taxonomy is applied consistently or exhaustively across the reviewed set.

    Authors: We agree that the classification requires explicit operational definitions and a mapping to ensure consistency. We will add clear definitions and decision rules for each bias category (observational, inductive, and learning bias) in the Results section, along with a table (or supplementary table) that maps each reviewed study to its assigned category with brief justification based on the paper's methodology. revision: yes

Circularity Check

0 steps flagged

No circularity: scoping review performs literature synthesis with no derivations or self-referential predictions

full rationale

This is a scoping review paper whose core activity is database search, classification of existing studies by problem type and ML strategy, and one external replication of a prior PINN implementation. No equations, fitted parameters, or predictions are defined within the paper that could reduce to its own inputs by construction. The 'often reaching accuracy comparable' statement is an aggregate observation drawn from the reviewed literature rather than a derived result internal to this manuscript. Self-citations, if present, are not load-bearing for any uniqueness theorem or ansatz. The work is self-contained as a descriptive synthesis against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The review's conclusions rest on the representativeness of the database search and the validity of the classification scheme; no free parameters or invented entities are introduced.

axioms (1)
  • domain assumption OpenAlex and Scopus databases provide sufficient coverage of relevant studies on physics-informed machine learning for wave propagation.
    Methods rely on these two databases for the literature search.

pith-pipeline@v0.9.1-grok · 5816 in / 1167 out tokens · 40088 ms · 2026-07-02T00:46:50.918157+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

92 extracted references · 62 canonical work pages · 3 internal anchors

  1. [1]

    The Leading Edge 13(9), 949–952 (1994) https://doi.org/10.1190/ 1.1437054

    Rocca, F., Wyatt, K.: SEG/EAEG 3-D mod- eling project: 2nd update. The Leading Edge 13(9), 949–952 (1994) https://doi.org/10.1190/ 1.1437054

    1994

  2. [2]

    Mechani- cal Systems and Signal Processing160(2021) https://doi.org/10.1016/j.ymssp.2021.107836

    Auriol, J., Kazemi, N., Niculescu, S.-I.: Sens- ing and computational frameworks for improv- ing drill-string dynamics estimation. Mechani- cal Systems and Signal Processing160(2021) https://doi.org/10.1016/j.ymssp.2021.107836

    work page doi:10.1016/j.ymssp.2021.107836 2021

  3. [3]

    Advances in Oil and Gas Exploration & Production

    Alsadi, H.N.: Seismic Hydrocarbon Explo- ration. Advances in Oil and Gas Exploration & Production. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-40436-3 . http://link.springer.com/10.1007/978-3-319- 40436-3

    work page doi:10.1007/978-3-319-40436-3 2017

  4. [4]

    Inter- national Journal of Social Research Method- ology8(1), 19–32 (2005) https://doi.org/10

    Arksey, H., O’Malley, L.: Scoping studies: towards a methodological framework. Inter- national Journal of Social Research Method- ology8(1), 19–32 (2005) https://doi.org/10. 1080/1364557032000119616

    2005

  5. [5]

    arXiv (2024)

    Alfarhan, M., Ravasi, M., Chen, F., Alkhalifah, T.: Robust Full Waveform Inversion with deep Hessian deblurring. arXiv (2024). https://doi. org/10.48550/arXiv.2403.17518

    work page doi:10.48550/arxiv.2403.17518 2024

  6. [6]

    Com- puter55(8), 40–48 (2022) https://doi.org/10

    Barba, L.A.: Defining the Role of Open Source Software in Research Reproducibility. Com- puter55(8), 40–48 (2022) https://doi.org/10. 1109/MC.2022.3177133

    work page arXiv 2022

  7. [7]

    Rand Corporation research study

    Bellman, R., Bellman, R.E., Corporation, R.: Dynamic Programming. Rand Corporation research study. Princeton University Press, ??? (1957)

    1957

  8. [8]

    GAMM-Mitteilungen 44(2), 202100006 (2021) https://doi.org/10

    Blechschmidt, J., Ernst, O.G.: Three ways to solve partial differential equations with neu- ral networks — A review. GAMM-Mitteilungen 44(2), 202100006 (2021) https://doi.org/10. 1002/gamm.202100006

    2021

  9. [9]

    Bezanson, J., Edelman, A., Karpinski, S., Shah, V.B.: Julia: A Fresh Approach to Numerical Computing (2014)

    2014

  10. [10]

    Cengage Learning, Boston, MA (2016)

    Burden, R.L., Faires, J.D., Burden, A.M.: Numer- ical Analysis. Cengage Learning, Boston, MA (2016)

    2016

  11. [11]

    http://github.com/jax-ml/ jax

    Wanderman-Milne, S., Zhang, Q.: JAX: com- posable transformations of Python+NumPy programs (2018). http://github.com/jax-ml/ jax

    2018

  12. [12]

    Bafghi, R.A., Raissi, M.: PINNs-Torch: Enhancing Speed and Usability of Physics-Informed Neural Networks with PyTorch. (2023)

    2023

  13. [13]

    Geophysics89(4), 331–345 (2024) https: //doi.org/10.1190/geo2023-0323.1

    Brandolin, F., Ravasi, M., Alkhalifah, T.: PINNs- lope: Seismic data interpolation and local slope estimation with physics informed neural net- works. Geophysics89(4), 331–345 (2024) https: //doi.org/10.1190/geo2023-0323.1

    work page doi:10.1190/geo2023-0323.1 2024

  14. [14]

    Journal of Scientific Computing92(3), 88 (2022) https://doi.org/10.1007/s10915-022-01939-z

    Cuomo, S., Di Cola, V.S., Giampaolo, F., Rozza, G., Raissi, M., Piccialli, F.: Scientific Machine Learning Through Physics–Informed Neural Networks: Where we are and What’s Next. Journal of Scientific Computing92(3), 88 (2022) https://doi.org/10.1007/s10915-022-01939-z

    work page doi:10.1007/s10915-022-01939-z 2022

  15. [15]

    Geophysics 85(1), 13–27 (2020) https://doi.org/10.1190/ geo2018-0562.1

    Cao, R., Earp, S., De Ridder, S.A.L., Cur- tis, A., Galetti, E.: Near-real-time near-surface 3D seismic velocity and uncertainty models by wavefield gradiometry and neural network inversion of ambient seismic noise. Geophysics 85(1), 13–27 (2020) https://doi.org/10.1190/ geo2018-0562.1

    2020

  16. [16]

    Campos, L.R., Nogueira, P., Moreira, D., Nasci- mento, E.G.S.: An empirical analysis of the influence of seismic data modeling for estimat- ing velocity models with fully convolutional networks, vol. 1, pp. 93–98 (2019) 27

    2019

  17. [17]

    Journal of Open Source Software5(46), 1931 (2020) https://doi.org/10.21105/joss.01931

    Chen, F., Sondak, D., Protopapas, P., Mattheakis, M., Liu, S., Agarwal, D., Giovanni, M.D.: Neu- roDiffEq: A Python package for solving differ- ential equations with neural networks. Journal of Open Source Software5(46), 1931 (2020) https://doi.org/10.21105/joss.01931

    work page doi:10.21105/joss.01931 1931

  18. [18]

    Geophysical Journal International215(2), 1257–1290 (2018) https://doi.org/10.1093/GJI/GGY283

    Beest, B., Goudswaard, J., Hohl, D.: Surro- gate regression modelling for fast seismogram generation and detection of microseismic events in heterogeneous velocity models. Geophysical Journal International215(2), 1257–1290 (2018) https://doi.org/10.1093/GJI/GGY283

    work page doi:10.1093/gji/ggy283 2018

  19. [19]

    Applied Math- ematical Modelling124, 325–352 (2023) https: //doi.org/10.1016/j.apm.2023.07.011

    Deng, W., Nguyen, K.T.P., Medjaher, K., Gogu, C., Morio, J.: Physics-informed machine learn- ing in prognostics and health management: State of the art and challenges. Applied Math- ematical Modelling124, 325–352 (2023) https: //doi.org/10.1016/j.apm.2023.07.011

    work page doi:10.1016/j.apm.2023.07.011 2023

  20. [20]

    Communications in Numerical Methods in Engineering10(3), 195–201 (1994) https://doi.org/10.1002/cnm

    Dissanayake, M.W.M.G., Phan-Thien, N.: Neural- network-based approximations for solving par- tial differential equations. Communications in Numerical Methods in Engineering10(3), 195–201 (1994) https://doi.org/10.1002/cnm. 1640100303

    work page doi:10.1002/cnm 1994

  21. [21]

    Drummond, D.C.: Replicability is not Repro- ducibility: Nor is it Good Science. (2009)

    2009

  22. [22]

    2022-August, pp

    Dhara, A., Sen, M.: Elastic-AdjointNet: A physics guided deep autoencoder to overcome cross talk effects in multiparameter full waveform inver- sion, vol. 2022-August, pp. 882–886 (2022). https://doi.org/10.1190/image2022-3745050.1 Escapil-Inchausp´ e, P., Ruz, G.A.: Hyper- parameter tuning of physics-informed neural networks: Application to Helmholtz pro...

    work page doi:10.1190/image2022-3745050.1 2022

  23. [23]

    Geoenergy Science and Engineering240(2024) https://doi.org/10.1016/j.geoen.2024.213028

    Fang, Z., Ba, J., Guo, Q., Xiong, F.: Shear-wave velocity prediction of tight reservoirs based on poroelasticity theory: A comparative study of deep neural network and rock physics model. Geoenergy Science and Engineering240(2024) https://doi.org/10.1016/j.geoen.2024.213028

    work page doi:10.1016/j.geoen.2024.213028 2024

  24. [24]

    arXiv (Cornell University) (2022) https: //doi.org/10.48550/arxiv.2204.13731

    Feng, Y., Chen, Y., Feng, S., Jin, P., Liu, Z., Lin, Y.: An Intriguing Property of Geophysics Inver- sion. arXiv (Cornell University) (2022) https: //doi.org/10.48550/arxiv.2204.13731

    work page doi:10.48550/arxiv.2204.13731 2022

  25. [25]

    1324 (2019)

    Fu, H., Zhang, Y., Ma, M.: Seismic wave- form inversion using a neural network-based forward, vol. 1324 (2019). https://doi.org/10. 1088/1742-6596/1324/1/012043

    2019

  26. [26]

    Sensors23(1) (2023) https://doi.org/ 10.3390/s23010061

    Gelboim, M., Adler, A., Sun, Y., Araya-Polo, M.: Encoder-Decoder Architecture for 3D Seismic Inversion. Sensors23(1) (2023) https://doi.org/ 10.3390/s23010061

    work page doi:10.3390/s23010061 2023

  27. [27]

    MIT Press, Cambridge, MA (2016)

    Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge, MA (2016)

    2016

  28. [28]

    International Journal of Hydrogen Energy61, 137–161 (2024) https://doi.org/10.1016/j.ijhydene.2024.02.296

    Gao, K., Creasy, N.M., Huang, L., Gross, M.R.: Underground hydrogen storage leakage detec- tion and characterization based on machine learning of sparse seismic data. International Journal of Hydrogen Energy61, 137–161 (2024) https://doi.org/10.1016/j.ijhydene.2024.02.296

    work page doi:10.1016/j.ijhydene.2024.02.296 2024

  29. [29]

    Journal of Seismology26(4), 781–821 (2022) https://doi.org/10.1007/s10950-021-10047-8

    Gosselin, J.M., Dosso, S.E., Askan, A., Wathelet, M., Savvaidis, A., Cassidy, J.F.: A review of inverse methods in seismic site characterization. Journal of Seismology26(4), 781–821 (2022) https://doi.org/10.1007/s10950-021-10047-8

    work page doi:10.1007/s10950-021-10047-8 2022

  30. [30]

    Grossmann, T.G., Komorowska, U.J., Latz, J., Sch¨ onlieb, C.-B.: Can physics-informed neural networks beat the finite element method? IMA Journal of Applied Mathematics89(1), 143– 174 (2024) https://doi.org/10.1093/imamat/ hxae011

    work page doi:10.1093/imamat/ 2024

  31. [31]

    Acta Geophysica 72(2), 673–687 (2024) https://doi.org/10.1007/ s11600-023-01123-3

    Guo, K., Zong, Z., Yang, J., Tan, Y.: Paramet- ric elastic full waveform inversion with con- volutional neural network. Acta Geophysica 72(2), 673–687 (2024) https://doi.org/10.1007/ s11600-023-01123-3

    2024

  32. [32]

    In: Advances in Subsurface Data Analytics: Traditional and Physics-Based Machine Learning, pp

    Huang, L., Clee, E., Ranasinghe, N.: Applying scientific machine learning to improve seismic wave simulation and inversion. In: Advances in Subsurface Data Analytics: Traditional and Physics-Based Machine Learning, pp. 167–192 (2022)

    2022

  33. [33]

    Hao, Z., Liu, S., Zhang, Y., Ying, C., Feng, Y., Su, H., Zhu, J.: Physics-Informed Machine 28 Learning: A Survey on Problems, Methods and Applications. arXiv. arXiv:2211.08064 [cs, math] (2023). http://arxiv.org/abs/2211.08064

    work page arXiv 2023

  34. [34]

    Neural Networks 4(2), 251–257 (1991) https://doi.org/10.1016/ 0893-6080(91)90009-T

    Hornik, K.: Approximation capabilities of mul- tilayer feedforward networks. Neural Networks 4(2), 251–257 (1991) https://doi.org/10.1016/ 0893-6080(91)90009-T

    1991

  35. [35]

    Computer Methods in Applied Mechanics and Engineering379, 113741 (2021) https://doi.org/10.1016/j.cma

    Juanes, R.: A physics-informed deep learning framework for inversion and surrogate mod- eling in solid mechanics. Computer Methods in Applied Mechanics and Engineering379, 113741 (2021) https://doi.org/10.1016/j.cma. 2021.113741

    work page doi:10.1016/j.cma 2021

  36. [37]

    Journal of Computational Physics409, 109313 (2020) https://doi.org/10.1016/j.jcp

    Hateley, J.C., Roberts, J., Mylonakis, K., Yang, X.: Deep learning seismic substructure detec- tion using the Frozen Gaussian approxima- tion. Journal of Computational Physics409, 109313 (2020) https://doi.org/10.1016/j.jcp. 2020.109313

    work page doi:10.1016/j.jcp 2020

  37. [38]

    Oxford University Press, Oxford (2017)

    Igel, H.: Computational Seismology: a Practical Introduction. Oxford University Press, Oxford (2017)

    2017

  38. [39]

    Progress in Geophysics 38(1), 430–448 (2023) https://doi.org/10.6038/ pg2023GG0142

    JingBo, Z.O.U., Cai, L.I.U., PengFei, Z.: Research progress of physics-informed neural network in seismic wave modeling. Progress in Geophysics 38(1), 430–448 (2023) https://doi.org/10.6038/ pg2023GG0142

    2023

  39. [40]

    IEEE Transactions on Geoscience and Remote Sensing62, 1–13 (2024) https://doi.org/10.1109/TGRS.2024.3409620

    Ji, D., Li, C., Zhai, C., Cao, Z.: An Efficient Platform for Numerical Modeling of Partial Differential Equations. IEEE Transactions on Geoscience and Remote Sensing62, 1–13 (2024) https://doi.org/10.1109/TGRS.2024.3409620

    work page doi:10.1109/tgrs.2024.3409620 2024

  40. [41]

    Nature Reviews Physics3(6), 422–440 (2021) https://doi.org/ 10.1038/s42254-021-00314-5

    Perdikaris, P., Wang, S., Yang, L.: Physics- informed machine learning. Nature Reviews Physics3(6), 422–440 (2021) https://doi.org/ 10.1038/s42254-021-00314-5 . Number: 6

    work page doi:10.1038/s42254-021-00314-5 2021

  41. [42]

    Geophysics72(2007) https://doi.org/10.1190/ 1.2757586

    Komatitsch, D., Martin, R.: An unsplit convo- lutional Perfectly Matched Layer improved at grazing incidence for the seismic wave equation. Geophysics72(2007) https://doi.org/10.1190/ 1.2757586

    2007

  42. [43]

    Geo- science Frontiers11(6), 1993–2001 (2020) https: //doi.org/10.1016/j.gsf.2020.07.007

    Karimpouli, S., Tahmasebi, P.: Physics informed machine learning: Seismic wave equation. Geo- science Frontiers11(6), 1993–2001 (2020) https: //doi.org/10.1016/j.gsf.2020.07.007

    work page doi:10.1016/j.gsf.2020.07.007 1993

  43. [44]

    Zenodo (2023)

    Nagaso, M., Rosenkrantz, E., Rusmanugroho, H., Andrade, E.S.d., Tape, C., Vilotte, J.- P., Xie, Z., Zhang, Z.: SPECFEM/specfem2d: SPECFEM2D v8.1.0. Zenodo (2023). https:// doi.org/10.5281/zenodo.10415228

    work page doi:10.5281/zenodo.10415228 2023

  44. [45]

    Zenodo (2024)

    Zhu, H.: SPECFEM/specfem3d: SPECFEM3D v4.1.1. Zenodo (2024). https://doi.org/10.5281/ zenodo.10823181

    2024

  45. [47]

    Nature521(7553), 436–444 (2015) https://doi

    LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature521(7553), 436–444 (2015) https://doi. org/10.1038/nature14539

    work page doi:10.1038/nature14539 2015

  46. [48]

    Proceedings of the Royal Society A: Math- ematical, Physical and Engineering Sciences 479(2275), 20230058 (2023) https://doi.org/10

    Lino, M., Fotiadis, S., Bharath, A.A., Cantwell, C.D.: Current and emerging deep-learning methods for the simulation of fluid dynamics. Proceedings of the Royal Society A: Math- ematical, Physical and Engineering Sciences 479(2275), 20230058 (2023) https://doi.org/10. 1098/rspa.2023.0058

    work page arXiv 2023

  47. [49]

    Com- puter Methods in Applied Mechanics and Engi- neering420, 116718 (2024) https://doi.org/10

    Lehmann, F., Gatti, F., Bertin, M., Clouteau, D.: Fourier Neural Operator Surrogate Model to Predict 3D Seismic Waves Propagation. Com- puter Methods in Applied Mechanics and Engi- neering420, 116718 (2024) https://doi.org/10. 1016/j.cma.2023.116718

    work page arXiv 2024

  48. [50]

    arXiv (2024)

    Lehmann, F., Gatti, F., Clouteau, D.: Multiple- Input Fourier Neural Operator (MIFNO) for source-dependent 3D elastodynamics. arXiv (2024). https://doi.org/10.48550/arXiv.2404. 10115

    work page doi:10.48550/arxiv.2404 2024

  49. [51]

    Li, Z., Kovachki, N., Azizzadenesheli, K., Liu, B., Bhattacharya, K., Stuart, A., Anandku- mar, A.: Neural Operator: Graph Kernel Net- work for Partial Differential Equations. arXiv. arXiv:2003.03485 [cs, math, stat] (2020). https: //doi.org/10.48550/arXiv.2003.03485

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2003.03485 2003

  50. [52]

    Fourier Neural Operator for Parametric Partial Differential Equations

    Bhattacharya, K., Stuart, A., Anandkumar, A.: Fourier Neural Operator for Parametric Par- tial Differential Equations. arXiv (2021). https: //doi.org/10.48550/arXiv.2010.08895

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2010.08895 2021

  51. [53]

    IEEE Transactions on Neural Networks9(5), 987–1000 (1998) https:// doi.org/10.1109/72.712178

    Lagaris, I.E., Likas, A., Fotiadis, D.I.: Artificial neural networks for solving ordinary and partial differential equations. IEEE Transactions on Neural Networks9(5), 987–1000 (1998) https:// doi.org/10.1109/72.712178 . Conference Name: IEEE Transactions on Neural Networks

    work page doi:10.1109/72.712178 1998

  52. [54]

    Geoscientific Model Development12(3), 1165–1187 (2019) https://doi.org/10.5194/gmd-12-1165-2019

    Gorman, G.J.: Devito (v3.1.0): an embed- ded domain-specific language for finite differ- ences and geophysical exploration. Geoscientific Model Development12(3), 1165–1187 (2019) https://doi.org/10.5194/gmd-12-1165-2019

    work page doi:10.5194/gmd-12-1165-2019 2019

  53. [55]

    Lu, L., Meng, X., Mao, Z., Karniadakis, G.E.: DeepXDE: A Deep Learning Library for Solving Differential Equations. SIAM Review 63(1), 208–228 (2021) https://doi.org/10.1137/ 19M1274067 L¨ ahivaara, T., Malehmir, A., Pasanen, A., K¨ arkk¨ ainen, L., Huttunen, J.M.J., Hesthaven, J.S.: Deep learning-based groundwater storage estimation from seismic data. (20...

    work page doi:10.3997/2214-4609.202229007 2021

  54. [56]

    Applied Sciences (Switzer- land)13(24) (2023) https://doi.org/10.3390/ app132413312

    Lu, C., Zhang, C.: Seismic Velocity Inversion via Physical Embedding Recurrent Neural Networks (RNN). Applied Sciences (Switzer- land)13(24) (2023) https://doi.org/10.3390/ app132413312

    2023

  55. [57]

    Mathematical Geosciences52(1), 53–79 (2020) https://doi.org/10.1007/s11004-019-09832-6

    Mosser, L., Dubrule, O., Blunt, M.J.: Stochastic Seismic Waveform Inversion Using Generative Adversarial Networks as a Geological Prior. Mathematical Geosciences52(1), 53–79 (2020) https://doi.org/10.1007/s11004-019-09832-6

    work page doi:10.1007/s11004-019-09832-6 2020

  56. [58]

    IEEE Transactions on Neural Networks and Learning Systems23(9), 1356– 1367 (2012) https://doi.org/10.1109/TNNLS

    Mehrkanoon, S., Falck, T., Suykens, J.A.K.: Approximate Solutions to Ordinary Differential Equations Using Least Squares Support Vec- tor Machines. IEEE Transactions on Neural Networks and Learning Systems23(9), 1356– 1367 (2012) https://doi.org/10.1109/TNNLS. 2012.2202126 . Conference Name: IEEE Trans- actions on Neural Networks and Learning Sys- tems

    work page doi:10.1109/tnnls 2012

  57. [59]

    Nature Machine Intelli- gence6(10), 1256–1269 (2024) https://doi.org/ 10.1038/s42256-024-00897-5

    McGreivy, N., Hakim, A.: Weak baselines and reporting biases lead to overoptimism in machine learning for fluid-related partial dif- ferential equations. Nature Machine Intelli- gence6(10), 1256–1269 (2024) https://doi.org/ 10.1038/s42256-024-00897-5

    work page doi:10.1038/s42256-024-00897-5 2024

  58. [60]

    Solid Earth 11(4), 1527–1549 (2020) https://doi.org/10

    Moseley, B., Nissen-Meyer, T., Markham, A.: Deep learning for fast simulation of seis- mic waves in complex media. Solid Earth 11(4), 1527–1549 (2020) https://doi.org/10. 5194/se-11-1527-2020 30

    2020

  59. [61]

    Advances in Geophysics48, 421–516 (2007) https://doi.org/10.1016/S0065-2687(06) 48008-0

    Moczo, P., Robertsson, J.O.A., Eisner, L.: The Finite-Difference Time-Domain Method for Modeling of Seismic Wave Propaga- tion. Advances in Geophysics48, 421–516 (2007) https://doi.org/10.1016/S0065-2687(06) 48008-0

    work page doi:10.1016/s0065-2687(06 2007

  60. [62]

    Neurocomputing159, 105–116 (2015) https://doi.org/10.1016/j.neucom.2015.02.013

    Mehrkanoon, S., Suykens, J.A.K.: Learning solu- tions to partial differential equations using LS- SVM. Neurocomputing159, 105–116 (2015) https://doi.org/10.1016/j.neucom.2015.02.013

    work page doi:10.1016/j.neucom.2015.02.013 2015

  61. [63]

    1038/d41586-020-02462-7

    Perkel, J.M.: Challenge to scientists: does your ten-year-old code still run? Nature 584(7822), 656–658 (2020) https://doi.org/10. 1038/d41586-020-02462-7

    2020

  62. [64]

    Exploration Geophysics55(3), 263–276 (2024) https://doi.org/10.1080/08123985.2024

    Park, Y., Moon, H.-J., Pyun, S.: Low-frequency marine seismic data reconstruction based on the far-field signature using a modified U- Net. Exploration Geophysics55(3), 263–276 (2024) https://doi.org/10.1080/08123985.2024. 2317129

    work page doi:10.1080/08123985.2024 2024

  63. [65]

    Neural Networks181, 106794 (2025).https://doi.org/https://doi.org/10.1016/j

    Roncoroni, G., Fortini, C., Bortolussi, L., Bienati, N., Pipan, M.: Synthetic seismic data genera- tion with deep learning. Journal of Applied Geo- physics190(2021) https://doi.org/10.1016/j. jappgeo.2021.104347 Rinc´ on, O.A., Perez Bernal, G., Montoya-

    work page doi:10.1016/j 2021

  64. [66]

    https://doi.org/ 10.5281/zenodo.21054377

    Noguera, S., Guar´ ın-Zapata, N.: Extracted Data from a Scoping Review of Machine Learning Approaches for Wave Propaga- tion Modeling in Seismology. https://doi.org/ 10.5281/zenodo.21054377 . https://doi.org/10. 5281/zenodo.21054377 Rinc´ on, O.A., Perez Bernal, G., Montoya-

    work page doi:10.5281/zenodo.21054377

  65. [67]

    https://doi.org/10.5281/zenodo.21017942

    Noguera, S., Guar´ ın-Zapata, N.: Literature Dataset of Identified Studies for a Scoping Review of Physics-Informed Machine Learning for Wave Propagation Modeling in Seismol- ogy. https://doi.org/10.5281/zenodo.21017942 . https://doi.org/10.5281/zenodo.21017942 Rinc´ on, O.A., Perez Bernal, G., Montoya-

    work page doi:10.5281/zenodo.21017942

  66. [68]

    Zenodo (2026)

    Noguera, S., Guar´ ın-Zapata, N.: Literature Dataset of Selected for a Scoping Review of Physics-Informed Machine Learning for Wave Propagation Mod- eling in Seismology. Zenodo (2026). https://doi.org/10.5281/zenodo.20834562 . https://doi.org/10.5281/zenodo.20834562

    work page doi:10.5281/zenodo.20834562 2026

  67. [69]

    URLhttps://www.sciencedirect

    Raissi, M., Perdikaris, P., Karniadakis, G.E.: Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial dif- ferential equations. Journal of Computational Physics378, 686–707 (2019) https://doi.org/ 10.1016/j.jcp.2018.10.045 R¨ oth, G., Tarantola, A.: Neural networks and inversion of ...

    work page doi:10.1016/j.jcp.2018.10.045 2019

  68. [70]

    IEEE Access8, 112266–112277 (2020) https://doi

    Ren, Y., Xu, X., Yang, S., Nie, L., Chen, Y.: A Physics-Based Neural-Network Way to Per- form Seismic Full Waveform Inversion. IEEE Access8, 112266–112277 (2020) https://doi. org/10.1109/ACCESS.2020.2997921

    work page doi:10.1109/access.2020.2997921 2020

  69. [71]

    Science 367(6481), 1026–1030 (2020) https://doi.org/ 10.1126/science.aaw4741

    Raissi, M., Yazdani, A., Karniadakis, G.E.: Hid- den fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science 367(6481), 1026–1030 (2020) https://doi.org/ 10.1126/science.aaw4741

    work page doi:10.1126/science.aaw4741 2020

  70. [72]

    Rivista del Nuovo Cimento43(9), 459–514 (2020) https: //doi.org/10.1007/s40766-020-00009-0

    Seriani, G., Oliveira, S.P.: Numerical modeling of mechanical wave propagation. Rivista del Nuovo Cimento43(9), 459–514 (2020) https: //doi.org/10.1007/s40766-020-00009-0

    work page doi:10.1007/s40766-020-00009-0 2020

  71. [73]

    Berlin (2009)

    Wiley-Blackwell, Malden, Mass. Berlin (2009)

    2009

  72. [74]

    Journal of Geophysics and Engineering 19(2), 269–282 (2022) https://doi.org/10.1093/ jge/gxac016

    Song, C., Wang, Y.: High-frequency wavefield extrapolation using the Fourier neural oper- ator. Journal of Geophysics and Engineering 19(2), 269–282 (2022) https://doi.org/10.1093/ jge/gxac016

    2022

  73. [75]

    Geophysical 31 Journal International232(3), 1503–1514 (2023) https://doi.org/10.1093/gji/ggac399

    Song, C., Wang, Y.: Simulating seismic multi- frequency wavefields with the Fourier feature physics-informed neural network. Geophysical 31 Journal International232(3), 1503–1514 (2023) https://doi.org/10.1093/gji/ggac399

    work page doi:10.1093/gji/ggac399 2023

  74. [76]

    SIAM, Philadelphia (2005)

    Tarantola, A.: Inverse Problem Theory and Meth- ods for Model Parameter Estimation. SIAM, Philadelphia (2005)

    2005

  75. [77]

    Annals of Internal Medicine169(7), 467–473 (2018) https://doi.org/10.7326/M18-0850

    Moriarty, J., Clifford, T., Tuncalp, O., Straus, S.E.: PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation. Annals of Internal Medicine169(7), 467–473 (2018) https://doi.org/10.7326/M18-0850

    work page doi:10.7326/m18-0850 2018

  76. [78]

    Results in Engineering13, 100316 (2022) https://doi.org/ 10.1016/j.rineng.2021.100316

    Matthews, E.: A review of physics-based machine learning in civil engineering. Results in Engineering13, 100316 (2022) https://doi.org/ 10.1016/j.rineng.2021.100316

    work page doi:10.1016/j.rineng.2021.100316 2022

  77. [79]

    Geophysical Prospecting 59(5), 794–813 (2011) https://doi.org/10.1111/ j.1365-2478.2011.00967.x

    Virieux, J., Calandra, H., Plessix, R.-E.: A review of the spectral, pseudo-spectral, finite-difference and finite-element modelling techniques for geophysical imaging. Geophysical Prospecting 59(5), 794–813 (2011) https://doi.org/10.1111/ j.1365-2478.2011.00967.x

    work page arXiv 2011

  78. [80]

    IEEE Transactions on Geoscience and Remote Sensing61(2023) https://doi.org/10.1109/TGRS.2023.3317529

    Wang, S., Jiang, Y., Song, P., Tan, J., Liu, Z., He, B.: Memory Optimization in RNN-Based Full Waveform Inversion Using Boundary Saving Wavefield Reconstruction. IEEE Transactions on Geoscience and Remote Sensing61(2023) https://doi.org/10.1109/TGRS.2023.3317529

    work page doi:10.1109/tgrs.2023.3317529 2023

  79. [81]

    2086–2090 (2018)

    Wang, W., Yang, F., Ma, J.: Velocity model build- ing with a modified fully convolutional network, pp. 2086–2090 (2018). https://doi.org/10.1190/ segam2018-2997566.1

    2086

  80. [82]

    2584–2588 (2019)

    Xu, Y., Li, J., Chen, X.: Physics informed neural networks for velocity inversion, pp. 2584–2588 (2019). https://doi.org/10.1190/ segam2019-3216823.1

    2019

Showing first 80 references.