fPINNs: Fractional Physics-Informed Neural Networks
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Abstract
Physics-informed neural networks (PINNs), introduced in [M. Raissi, P. Perdikaris, and G. Karniadakis, J. Comput. Phys., 378 (2019), pp. 686--707], are effective in solving integer-order partial differential equations (PDEs) based on scattered and noisy data. PINNs employ standard feedforward neural networks (NNs) with the PDEs explicitly encoded into the NN using automatic differentiation, while the sum of the mean-squared PDE residuals and the mean-squared error in initial-boundary conditions is minimized with respect to the NN parameters. Here we extend PINNs to fractional PINNs (fPINNs) to solve space-time fractional advection-diffusion equations (fractional ADEs), and we study systematically their convergence, hence explaining both fPINNs and PINNs for the first time. Specifically, we demonstrate their accuracy and effectiveness in solving multidimensional forward and inverse problems with forcing terms whose values are only known at randomly scattered spatio-temporal coordinates ( black-box (BB) forcing terms). A novel element of the fPINNs is the hybrid approach that we introduce for constructing the residual in the loss function using both automatic differentiation for the integer-order operators and numerical discretization for the fractional operators. This approach bypasses the difficulties stemming from the fact that automatic differentiation is not applicable to fractional operators because the standard chain rule in integer calculus is not valid in fractional calculus. To discretize the fractional operators, we employ the Grünwald--Letnikov (GL) formula in one-dimensional fractional ADEs and the vector GL formula in conjunction with the directional fractional Laplacian in two- and three-dimensional fractional ADEs. We first consider the one-dimensional fractional Poisson equation and compare the convergence of the fPINNs against the finite difference method (FDM). We present the solution convergence using both the mean $L^2$ error as well as the standard deviation due to sensitivity to NN parameter initializations. Using different GL formulas we observe first-, second-, and third-order convergence rates for small size training sets but the error saturates for larger training sets. We explain these results by analyzing the four sources of numerical errors due to discretization, sampling, NN approximation, and optimization. The total error decays monotonically (below $10^{-5}$ for a third-order GL formula) but it saturates beyond that point due to the optimization error. We also analyze the relative balance between discretization and sampling errors and observe that the sampling size and the number of discretization points ( auxiliary points) should be comparable to achieve the highest accuracy. As we increase the depth of the NN up to certain value, the mean error decreases and the standard deviation increases, whereas the width has essentially no effect unless its value is either too small or too large. We next consider time-dependent fractional ADEs and compare white-box (WB) and BB forcing. We observe that for the WB forcing, our results are similar to the aforementioned cases; however, for the BB forcing fPINNs outperform FDM. Subsequently, we consider multidimensional time-, space-, and space-time-fractional ADEs using the directional fractional Laplacian and we observe relative errors of $10^{-3}\sim10^{-4}$. Finally, we solve several inverse problems in one, two, and three dimensions to identify the fractional orders, diffusion coefficients, and transport velocities and obtain accurate results given proper initializations even in the presence of significant noise.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“We also used the extended stochastic gradient descent Adam algorithm [18] to optimize the loss function.”
MethodRaissi, M., Perdikaris, P., & Karniadakis, G. E. (2019) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.10.045 · Cited by 20,384 · cited 4×
“We just need to add the parameters to be identified in the inverse problem to the list of parameters to be optimized in the forward problem without changing anything else [35].”
BackgroundMethodInfluentialSelf-citationModel-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
Finn, C., Abbeel, P., & Levine, S. (2017) · arXiv · doi:10.48550/arxiv.1703.03400 · Cited by 8,512 · cited 1×
“We can tune these attributes of NN architecture automatically by leveraging metalearning techniques [65, 66].”
MethodImproving neural networks by preventing co-adaptation of feature detectors
Hinton, G. E., Srivastava, N., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. R. (2012) · arXiv · doi:10.48550/arxiv.1207.0580 · Cited by 6,382 · cited 1×
“Additionally, the noise in the data can be naturally taken into account by employing regularization techniques, such as L(1), L(2) and L∞ regularization [32], early stopping [33], as well as dropout [34, 35].”
MethodDiscovering governing equations from data by sparse identification of nonlinear dynamical systems
Brunton, S. L., Proctor, J. L., & Kutz, J. N. (2016) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1517384113 · Cited by 4,796 · cited 10×
“In this paper, we focus on the identification of the parameters of fractional PDEs whose overall form is known but some coefficients and most importantly the operators themselves are not known.”
BackgroundInfluentialNeural Architecture Search with Reinforcement Learning
Zoph, B., & Le, Q. V. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.01578 · Cited by 3,974 · cited 1×
“We can tune these attributes of NN architecture automatically by leveraging meta-learning techniques [54].”
MethodArtificial neural networks for solving ordinary and partial differential equations
Lagaris, I. E., Likas, A., & Fotiadis, D. I. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/72.712178 · Cited by 2,412 · cited 1×
“In particular, we concentrate on physics-informed neural networks (PINNs) [7, 46, 19, 35], which belong to the second aforementioned category.”
BackgroundHidden physics models: Machine learning of nonlinear partial differential equations
Raissi, M., & Karniadakis, G. E. (2018) · Journal of Computational Physics · doi:10.1016/j.jcp.2017.11.039 · Cited by 1,349 · cited 1×
“Other examples include, but not limited to, [21, 22, 23].”
BackgroundSelf-citationApplication of a fractional advection‐dispersion equation
Benson, D. A., Wheatcraft, S. W., & Meerschaert, M. M. (2000) · Water Resources Research · doi:10.1029/2000wr900031 · Cited by 1,301 · cited 1×
“Among these are solute transport in fractured and porous media [2, 3], acoustic wave propagation with frequency-dependent dissipation [4, 5, 6, 7], laminar and turbulent flows [8, 9, 10], viscoelastic constitutive laws [11], just to mention a few.”
BackgroundA fully discrete difference scheme for a diffusion-wave system
Sun, Z.-Z., & Wu, X. (2006) · Applied Numerical Mathematics · doi:10.1016/j.apnum.2005.03.003 · Cited by 1,235 · cited 2×
“The truncating error for the L1 scheme is (\Delta t) 2 - \gamma [44].”
Method
Research works printed but not linked: 31
- [62] Avrim Blum and Ronald L Rivest. Training a 3-node neural network is np-complete. In Advances in neural information processing systems , pages 494-501, 1989. -
- [13] Samuel H Rudy, Steven L Brunton, Joshua L Proctor, and J Nathan Kutz. Data-driven discovery of partial differential equations. Science Advances , 3(4):e1602614, 2017. -
- [14] Zichao Long, Yiping Lu, Xianzhong Ma, and Bin Dong. PDE-net: Learning PDEs from data. In Jennifer Dy and Andreas Krause, editors, Proceedings of the 35th International Conference on Machine Learning , volume 80 of Proceedings of Machine Learning Research , pages 32083216, Stockholmsmässan, Stockholm Sweden, 10-15 Jul 2018. PMLR. -
- [11] Francesco Mainardi. Fractional calculus and waves in linear viscoelasticity: an introduction to mathematical models . World Scientific, 2010. -
- [5] Bradley E Treeby and BT Cox. Modeling power law absorption and dispersion for acoustic propagation using the fractional laplacian. The Journal of the Acoustical Society of America , 127(5):2741-2748, 2010. -
- 10.1007/978-3-642-14574-2 · doi:10.1007/978-3-642-14574-2
- [57] Rudolf Gorenflo, Anatoli˘ ı Aleksandrovich Kilbas, Francesco Mainardi, Sergei V Rogosin, et al. Mittag-Leffler functions, related topics and applications , volume 2. Springer, 2014. -
- [19] Vladimir I Gorbachenko, Tatiana V Lazovskaya, Dmitriy A Tarkhov, Alexander N Vasilyev, and Maxim V Zhukov. Neural network technique in some inverse problems of mathematical physics. In International Symposium on Neural Networks , pages 310-316. Springer, 2016. -
- [59] Xavier Glorot and Yoshua Bengio. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the thirteenth international conference on artificial intelligence and statistics , pages 249-256, 2010. -
- 10.1007/978-3-319-40663-3_36 · doi:10.1007/978-3-319-40663-3_36
- 10.1007/978-3-662-43930-2 · doi:10.1007/978-3-662-43930-2
- [47] A Anatolii Aleksandrovich Kilbas, Hari Mohan Srivastava, and Juan J Trujillo. Theory and applications of fractional differential equations , volume 204. Elsevier Science Limited, 2006. -
- [23] Mamikon Gulian, Maziar Raissi, Paris Perdikaris, and George Karniadakis. Machine learning of space-fractional differential equations. arXiv preprint arXiv:1808.00931 , 2018. - · arXiv
- [24] Tianping Chen and Hong Chen. Approximations of continuous functionals by neural networks with application to dynamic systems. IEEE Transactions on Neural Networks , 4(6):910-918, 1993. -
- [7] Tieyuan Zhu and Jerry M Harris. Modeling acoustic wave propagation in heterogeneous attenuating media using decoupled fractional laplacians. Geophysics , 79(3):T105-T116, 2014. -
- [54] Bangti Jin, Raytcho Lazarov, and Zhi Zhou. An analysis of the l1 scheme for the subdiffusion equation with nonsmooth data. IMA Journal of Numerical Analysis , 36(1):197-221, 2015. -
- Kilbas A. A. A., 2006, North-Holland Math. Stud. 204
- [61] Jason D Lee, Max Simchowitz, Michael I Jordan, and Benjamin Recht. Gradient descent only converges to minimizers. In Conference on Learning Theory , pages 1246-1257, 2016. -
- [67] Lihua Lei, Cheng Ju, Jianbo Chen, and Michael I Jordan. Non-convex finite-sum optimization via scsg methods. In Advances in Neural Information Processing Systems , pages 2348-2358, 2017. -
- [68] Chenxin Ma, Jakub Konečn` y, Martin Jaggi, Virginia Smith, Michael I Jordan, Peter Richtárik, and Martin Takáč. Distributed optimization with arbitrary local solvers. Optimization Methods and Software , 32(4):813-848, 2017.
- 10.1142/p614 · doi:10.1142/p614
- Meerschaert M. M., 2004, Fract. Calc. Appl. Anal., 7, 61
- [41] Ying Zhang and Xiang Xu. Inverse source problem for a fractional diffusion equation. Inverse problems , 27(3):035010, 2011. -
- [42] Bangti Jin and William Rundell. An inverse problem for a one-dimensional time-fractional diffusion problem. Inverse Problems , 28(7):075010, 2012. -
- [50] Igor Podlubny. Fractional differential equations: an introduction to fractional derivatives, fractional differential equations, to methods of their solution and some of their applications. Mathematics in science and engineering , 198:261-300, 1999. -
- [51] Kai Diethelm. The analysis of fractional differential equations: An application-oriented exposition using differential operators of Caputo type . Springer Science & Business Media, 2010. -
- [32] Yoshua Bengio, Ian J Goodfellow, and Aaron Courville. Deep learning. Nature , 521(7553):436444, 2015. -
- Samko S. G., 1993, Gordon and Breach
- [64] John Paul Roop. Computational aspects of fem approximation of fractional advection dispersion equations on bounded domains in r2. Journal of Computational and Applied Mathematics , 193(1):243-268, 2006. -
- [9] Vladimir V Uchaikin. Fractional derivatives for physicists and engineers , volume 2. Springer, 2013. -
- [35] Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. Dropout: a simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research , 15(1):1929-1958, 2014. -
Cited by
Items 201–300 of 1,000
Masclans, N., Vázquez-Novoa, F., Bernades, M., Badia, R. M., & Jofre, L. (2023) · International Journal of Thermofluids · doi:10.1016/j.ijft.2023.100448 · Cited by 40
Wang, S., & Karniadakis, G. E. (2024) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2024.117189 · Cited by 36
Self-citationYuan, B., Wang, H., Heitor, A., & Chen, X. (2024) · Journal of Computational Physics · doi:10.1016/j.jcp.2024.113284 · Cited by 36
Review of machine learning applications for defect detection in composite materials
Daghigh, V., Daghigh, H., Lacy, T. E., & Naraghi, M. (2024) · Machine Learning with Applications · doi:10.1016/j.mlwa.2024.100600 · Cited by 46
Physics-constrained machine learning for chemical engineering
Mukherjee, A., & Zavala, V. M. (2026) · Current Opinion in Chemical Engineering · doi:10.1016/j.coche.2026.101228 · Cited by 42
Zhang, Z., Zhang, H., Liu, Y., Li, J.-Y., & Liu, C. (2023) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2023.113169 · Cited by 38
Fully probabilistic deep models for forward and inverse problems in parametric PDEs
Vadeboncoeur, A., Akyıldız, Ö. D., Kazlauskaite, I., Girolami, M., & Cirak, F. (2023) · Journal of Computational Physics · doi:10.1016/j.jcp.2023.112369 · Cited by 42
“PINN-based methods [52, 53, 47, 45, 33, 9] convert the PDE formulation into a loss function with either hard or soft constraints [44, 70].”
MethodDeep Hamiltonian networks based on symplectic integratorsPreprint
Zhu, A., Jin, P., & Tang, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.13830 · Cited by 38
“In particular, neural networks have been applied to solve differential equations [11, 27, 34, 42].”
MethodMa, X., Qiu, L., Zhang, B., Wu, G., & Wang, F. (2024) · International Journal of Heat and Mass Transfer · doi:10.1016/j.ijheatmasstransfer.2024.126393 · Cited by 38
Liu, Y., Gu, H., Yu, X., & Qin, P. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106886 · Cited by 46
Cai, K., & Wang, J. (2024) · Physics of Fluids · doi:10.1063/5.0244094 · Cited by 46
Jiao, A., Yan, Q., Harlim, J., & Lu, L. (2026) · SIAM Journal on Scientific Computing · doi:10.1137/24m1675254 · Cited by 39
Self-citationThe deep parametric PDE method and applications to option pricing
Glau, K., & Wunderlich, L. (2022) · Applied Mathematics and Computation · doi:10.1016/j.amc.2022.127355 · Cited by 35
Wu, S., Zhu, A., Tang, Y., & Lu, B. (2023) · Communications in Computational Physics · doi:10.4208/cicp.oa-2022-0218 · Cited by 35
Hu, Z., Kawaguchi, K., Zhang, Z., & Karniadakis, G. E. (2024) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2024.117448 · Cited by 34
Self-citationAlaïa, E. B., Dhahri, S., & Naifar, O. (2025) · IEEE Access · doi:10.1109/access.2025.3595958 · Cited by 34
“Our kernel selection methodology draws inspiration from Rossikhin’s solid mechanics applications [29] and Sun’s multidisciplinary survey [30], which validate the physical interpretability of kernel functions.”
Inferring vortex induced vibrations of flexible cylinders using physics-informed neural networks
Kharazmi, E., Fan, D., Wang, Z., & Triantafyllou, M. S. (2021) · Journal of Fluids and Structures · doi:10.1016/j.jfluidstructs.2021.103367 · Cited by 34
Pu, J., & Chen, Y. (2022) · Communications in Nonlinear Science and Numerical Simulation · doi:10.1016/j.cnsns.2022.107051 · Cited by 38
PMNN: Physical model-driven neural network for solving time-fractional differential equations
Ma, Z., Hou, J., Zhu, W., Peng, Y., & Li, Y. (2023) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2023.114238 · Cited by 33
Lin, S., & Chen, Y. (2023) · Physica D Nonlinear Phenomena · doi:10.1016/j.physd.2023.134023 · Cited by 33
Machine learning prediction of materials properties from chemical composition: Status and prospects
Alghadeer, M., Aisyah, N. D., Hezam, M., Alqahtani, S. M., Baloch, A. A. B., & Alharbi, F. H. (2024) · Chemical Physics Reviews · doi:10.1063/5.0235541 · Cited by 37
Data-driven discovery of coarse-grained equations
Bakarji, J., & Tartakovsky, D. M. (2021) · Journal of Computational Physics · doi:10.1016/j.jcp.2021.110219 · Cited by 41
Physics-informed surrogate modeling for hydro-fracture geometry prediction based on deep learning
Lu, Y., Wang, B., Zhao, Y., Yang, X., Li, L., Dong, M., Lv, Q., Zhou, F., Gu, N., & Shang, L. (2022) · Energy · doi:10.1016/j.energy.2022.124139 · Cited by 33
“Guo et al. [18] extended PINNs to solve space-time fractional advection-diffusion equations.”
BackgroundWang, T., Wang, Z., Huang, Z., & Xi, G. (2022) · Science China Technological Sciences · doi:10.1007/s11431-022-2118-9 · Cited by 34
Mostajeran, F., & Hosseini, S. M. (2023) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2023.04.026 · Cited by 33
Data-driven learning of nonlocal models: from high-fidelity simulations to constitutive laws
Sandia National Laboratories (SNL-CA), L., You, H., (NNSA), U. N. N. S. A., Sandia National Laboratories, A., Yu, Y., Silling, S., & D’Elia, M. (2020) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · doi:10.2172/1735693 · Cited by 33
Zhou, M., & Mei, G. (2023) · Mathematics · doi:10.3390/math11112529 · Cited by 32
“Several PINN variants have emerged to address different problems, including conservation PINNs (cPINNs) [52], variational PINNs (vPINNs) [53], fractional-order PINNs (fPINNs) [54], and others.”
BackgroundDeep fuzzy physics-informed neural networks for forward and inverse PDE problems
Wu, W., Duan, S., Sun, Y., Yang, Y., Liu, D., & Peng, D. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106750 · Cited by 32
CMINNs: Compartment model informed neural networks — Unlocking drug dynamics
Daryakenari, N. A., Wang, S., & Karniadakis, G. E. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.109392 · Cited by 42
Self-citationMarelli, S., Wagner, P., Lataniotis, C., & Sudret, B. (2020) · International Journal for Uncertainty Quantification · doi:10.1615/int.j.uncertaintyquantification.2020034395 · Cited by 32
“Physics-informed neural networks are reaching unprecedented approximation power in UQ applications (see, e.g. , (Raissi et al., 2019; Pang et al., 2019)), while sparse polynomial chaos expansions are used as denoising regressors in Torre et al. (2019a), as high-dimensional regression tools in…”
Fractional Chebyshev deep neural network (FCDNN) for solving differential models
Hajimohammadi, Z., Baharifard, F., Ghodsi, A., & Parand, K. (2021) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2021.111530 · Cited by 37
Darbon, J., Dower, P. M., & Meng, T. (2022) · Mathematics of Control Signals and Systems · doi:10.1007/s00498-022-00333-2 · Cited by 41
“The related works include but are not limited to [4,8,12,15–17,19,26,30,41–43,51,53,59,63,68,70,71,77,79,80,83,91, 95,96,102–104,107,111–113,115,121–125,128,130,132–138,140,142,144,146, 147,149,151–153,157,158].”
Physics-informed neural networks for incompressible flows with moving boundaries
Zhu, Y., Kong, W., Deng, J., & Bian, X. (2024) · Physics of Fluids · doi:10.1063/5.0186809 · Cited by 42
Patawari, A., & Das, P. (2025) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-025-10343-3 · Cited by 31
Huang, Y., Hao, W., & Lin, G. (2022) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2022.07.002 · Cited by 39
Zhou, Z., Wang, L., & Yan, Z. (2022) · Physics Letters A · doi:10.1016/j.physleta.2022.128373 · Cited by 33
“The powerful physics-informed neural network (PINN) method [38,39] was used to investigate the PDEs [40–48], the fractional PDEs [49], and stochastic PDEs [50].”
MethodZhao, F., & Hung, D. L. S. (2022) · Applied Thermal Engineering · doi:10.1016/j.applthermaleng.2022.119633 · Cited by 34
You, H., Yu, Y., Silling, S. A., & D’Elia, M. (2024) · Journal of Peridynamics and Nonlocal Modeling · doi:10.1007/s42102-024-00119-x · Cited by 30
“The literature on techniques for learning kernel parameters for a given functional form is vast, see, e.g., [14, 15] for control-based approaches, and [16, 17] for machine-learning approaches.”
Physics-informed neural fractional differential equations
Vellappandi, M., & Lee, S. (2025) · Applied Mathematical Modelling · doi:10.1016/j.apm.2025.116127 · Cited by 29
PHYSICS-INFORMED NEURAL NETWORK FOR SOLVING HAUSDORFF DERIVATIVE POISSON EQUATIONS
Wu, G., Wang, F., & Qiu, L. (2023) · Fractals · doi:10.1142/s0218348x23401035 · Cited by 29
Li, X., Liu, Y., & Liu, Z. (2023) · Physics of Fluids · doi:10.1063/5.0151244 · Cited by 45
A deep neural network approach for parameterized PDEs and Bayesian inverse problems
Antil, H., Elman, H. C., Onwunta, A., & Verma, D. (2023) · Machine Learning Science and Technology · doi:10.1088/2632-2153/ace67c · Cited by 32
Bhaumik, B., De, S., & Changdar, S. (2023) · Mathematics and Computers in Simulation · doi:10.1016/j.matcom.2023.10.011 · Cited by 34
Encoding Invariances in Deep Generative ModelsPreprint
Shah, V., Joshi, A., Ghosal, S., Pokuri, B. S. S., Sarkar, S., Ganapathysubramanian, B., & Hegde, C. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.01626 · Cited by 29
“Pang et al. [18] extend this to fractional PDEs whle providing theoretical guarantees for convergence.”
BackgroundPhysics-informed neural networks for solving inverse problems in phase field models
Zhao, B.-W., Sun, D., Wu, H., Qin, C., & Fei, Q. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107665 · Cited by 28
Hu, Z., Yang, Z., Wang, Y., Karniadakis, G. E., & Kawaguchi, K. (2025) · SIAM Journal on Scientific Computing · doi:10.1137/23m1621356 · Cited by 28
Self-citationMamun, K. M., Pala, N., & Shawkat, M. S. A. (2025) · IEEE Access · doi:10.1109/access.2025.3605856 · Cited by 32
“The architecture of a PINN is similar to that of an artificial neural network (ANN), consisting of multiple layers of neurons [73].”
Magnetohydrodynamics with physics informed neural operators
Rosofsky, S., & Huerta, E. A. (2023) · Machine Learning Science and Technology · doi:10.1088/2632-2153/ace30a · Cited by 28
Ma, L., Li, R. L. R., Zeng, F., Guo, L. G. L., & Karniadakis, G. E. (2023) · Communications in Computational Physics · doi:10.4208/cicp.oa-2023-0075 · Cited by 40
Self-citation郭, J. 嘉. G., 王, H. 海. W., 古, S. 仕. G., & Hou, C. (2024) · Chinese Physics B · doi:10.1088/1674-1056/ad21f3 · Cited by 36
HNS: An efficient hermite neural solver for solving time-fractional partial differential equations
Hou, J., Ma, Z., Ying, S., & Li, Y. (2024) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2024.114637 · Cited by 27
Wang, N., Chen, Y., & Zhang, D. (2025) · The Innovation Energy · doi:10.59717/j.xinn-energy.2025.100087 · Cited by 36
Eshkofti, K., & Hosseini, S. M. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106908 · Cited by 27
Faroughi, S., Darvishi, A., & Rezaei, S. (2023) · Acta Mechanica · doi:10.1007/s00707-023-03676-2 · Cited by 29
“This method is often extended to obtain unique paths for dealing with related mathematical formulas, such as stochastic PDEs [42] and fractional PDEs [43].”
MethodPhysics-informed neural networks (P INNs): application categories, trends and impact
Ghalambaz, M., Sheremet, M. А., Khan, M. A., Raizah, Z. A. S., & Shafi, J. (2024) · International Journal of Numerical Methods for Heat & Fluid Flow · doi:10.1108/hff-09-2023-0568 · Cited by 27
A Unified Theory of Fractional Nonlocal and Weighted Nonlocal Vector Calculus
Sandia National Laboratories (SNL-NM), A., D’Elia, M., (NNSA), U. N. N. S. A., Gulian, M., Olson, H., & Karniadakis, G. E. (2020) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · doi:10.2172/1618398 · Cited by 27
“, volume or exterior conditions; the unresolved treatment of nonlocal interfaces [5, 13]; the uncertainty and sparsity of model parameters and data [7, 21, 22, 36, 50, 51], and the increasing computational cost as the extent of the nonlocal interactions increases [3, 4, 18, 23, 52, 65, 68].”
BackgroundSelf-citationYu, X., & Zhou, X. (2023) · Computational Mechanics · doi:10.1007/s00466-023-02365-0 · Cited by 26
Gomez-Flores, A., Cho, H., Hong, G., Nam, H., Kim, H., & Chung, Y.-S. (2024) · Materials & Design · doi:10.1016/j.matdes.2024.113247 · Cited by 26
Fractional deep neural network via constrained optimization
Antil, H., Khatri, R., Löhner, R., & Verma, D. (2020) · Machine Learning Science and Technology · doi:10.1088/2632-2153/aba8e7 · Cited by 39
Solving 1D non‐linear magneto quasi‐static Maxwell's equations using neural networks
Baldan, M., Baldan, G., & Nacke, B. (2021) · IET Science Measurement & Technology · doi:10.1049/smt2.12022 · Cited by 26
“Furthermore, PINNs were extended to fractional PINNs (fPINNs) to solve fractional advection-diffusion equations [11] and, based on this paradigm, Lu et al. [10] developed the Deep-XDE library, capable of solving a wide range of differential equations.”
MethodBora, A., Dai, W., Wilson, J. P., & Boyt, J. C. (2021) · International Journal of Heat and Mass Transfer · doi:10.1016/j.ijheatmasstransfer.2021.121616 · Cited by 31
NeuFENet: neural finite element solutions with theoretical bounds for parametric PDEs
Khara, B., Balu, A., Joshi, A., Sarkar, S., Hegde, C., Krishnamurthy, A., & Ganapathysubramanian, B. (2024) · Engineering With Computers · doi:10.1007/s00366-024-01955-7 · Cited by 28
A deep learning approach for the solution of probability density evolution of stochastic systems
Pourtakdoust, S. H., & Khodabakhsh, A. H. (2022) · Structural Safety · doi:10.1016/j.strusafe.2022.102256 · Cited by 26
“Furthermore, AD can also be employed recursively to compute the n-order derivative of the network as well, and again, since the derivation is already taken care of with AD, noisy data and other similar numerical instabilities do not raise an issue [26].”
BackgroundA physics-constrained neural network for multiphase flows
Zheng, H., Huang, Z., & Lin, G. (2022) · Physics of Fluids · doi:10.1063/5.0111275 · Cited by 26
“[40] developed fractional physics-informed neural networks (fPINNs) to encode fractional-order partial differential equations (PDEs).”
BackgroundWang, X., Zhu, S., Guo, Y., Han, P., Wang, Y., Wei, Z., & Jin, X. (2022) · Journal of Computational Science · doi:10.1016/j.jocs.2022.101906 · Cited by 28
Sensitivity analysis using Physics-informed neural networks
Hanna, J. M., Aguado, J. V., Comas-Cardona, S., Askri, R., & Borzacchiello, D. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.108764 · Cited by 26
Transparency: The Missing Link to Boosting AI Transformations in Chemical Engineering
Yuan, Y., Chaffart, D., Wu, T., & Zhu, J. (2024) · Engineering · doi:10.1016/j.eng.2023.11.024 · Cited by 25
Ghoreyshi, A., Abbaszadeh, M., Zaky, M. A., & Dehghan, M. (2025) · Applied Numerical Mathematics · doi:10.1016/j.apnum.2025.03.002 · Cited by 25
Liu, C., & Wu, H. A. (2023) · International Journal of Applied Mechanics · doi:10.1142/s1758825123500655 · Cited by 25
DMIS: Dynamic Mesh-Based Importance Sampling for Training Physics-Informed Neural Networks
Yang, Z., Qiu, Z., & Fu, D. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i4.25669 · Cited by 25
Multi-objective inverse design of finned heat sink system with physics-informed neural networks
Lu, Z., Li, Y., He, C., Ren, J., Yu, H., Zhang, B., & Chen, Q. (2023) · Computers & Chemical Engineering · doi:10.1016/j.compchemeng.2023.108500 · Cited by 25
Rani, N., & Fernandez, A. (2024) · Communications in Nonlinear Science and Numerical Simulation · doi:10.1016/j.cnsns.2024.108249 · Cited by 24
Pre-trained Physics-Informed Neural Networks for Analysis of Contaminant Transport in Soils
Ke, Z.-W., Wei, S.-J., Yao, S.-Y., Chen, S., Chen, Y., & Li, Y.-C. (2025) · Computers and Geotechnics · doi:10.1016/j.compgeo.2025.107055 · Cited by 24
Neural fractional differential equations
Coelho, C., Costa, M. F. P., & Ferrás, L. L. (2025) · Applied Mathematical Modelling · doi:10.1016/j.apm.2025.116060 · Cited by 24
“It should be remarked that Neural ODEs and Neural FDEs are often confused with other methods in the literature that use NNs to approximate solutions of ODEs or FDEs [4, 5], where the ODE or FDE modelling the data dynamics is already known.”
Physics-informed neural networks for modeling turbulent combustion
Cao, Z., Luo, K., Liu, K., Cheng, Y., Xing, J., Jiang, L., & Fan, J. (2025) · Fuel · doi:10.1016/j.fuel.2025.136475 · Cited by 24
Baniya, S., & Maity, D. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11444-y · Cited by 26
Error estimates for deep learning methods in fluid dynamics
Biswas, A., Tian, J., & Ulusoy, S. (2022) · Numerische Mathematik · doi:10.1007/s00211-022-01294-z · Cited by 24
“Moreover, machine learning methods (physics informed neural networks [21, 22, 25, 27, 38] which are mesh-free) have gained wide applicability in obtaining numerical solutions of various types of partial differential equations (PDEs); see [2, 3, 12, 21, 23, 26–28,”
MethodWang, K., & Si, J. (2022) · Mathematical Methods in the Applied Sciences · doi:10.1002/mma.8588 · Cited by 24
BINet: Learn to Solve Partial Differential Equations with Boundary Integral Networks
Lin, G., Hu, P., Chen, F., Chen, X. C. X., Chen, J. C. J., Wang, J., & Shi, Z. S. Z. (2023) · CSIAM Transactions on Applied Mathematics · doi:10.4208/csiam-am.so-2022-0014 · Cited by 30
Deep learning approximations for non-local nonlinear PDEs with Neumann boundary conditions
Boussange, V., Becker, S. B. S., Jentzen, A., Kuckuck, B., & Pellissier, L. (2023) · Partial Differential Equations and Applications · doi:10.1007/s42985-023-00244-0 · Cited by 24
“While some of these approaches use classical methods susceptible to the curse of dimensionality for the non-local part [81, 72], mesh-free methods suitable for high-dimensional problems have also been investigated [47, 2, 94].”
MethodThree-dimensional laminar flow using physics informed deep neural networks
Biswas, S. K., & Anand, N. K. (2023) · Physics of Fluids · doi:10.1063/5.0180834 · Cited by 24
Wang, X., Wu, Z., Song, J. M., Han, W., & Yan, Z. (2024) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2024.114509 · Cited by 24
I-FENN for thermoelasticity based on physics-informed temporal convolutional network (PI-TCN)
Abueidda, D., & Mobasher, M. E. (2024) · Computational Mechanics · doi:10.1007/s00466-024-02475-3 · Cited by 24
“Additionally, several variations of PINN models have been proposed in an attempt to improve the performance of PINN models, such as h-PINNs [35], VPINNs [36], fPINNs [37], and B-PINNs [38].”
BackgroundSandia National Laboratories (SNL-NM), A., Pang, G., (NNSA), U. N. N. S. A., Brown Univ., P., D’Elia, M., Parks, M., & Karniadakis, G. E. (2020) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · doi:10.2172/1614899 · Cited by 24
Self-citationBayesian neural networks for weak solution of PDEs with uncertainty quantificationPreprint
Zhang, X., & Garikipati, K. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.04879 · Cited by 26
An enhanced hybrid adaptive physics-informed neural network for forward and inverse PDE problems
Luo, K., Liao, S., Guan, Z., & Liu, B. (2025) · Applied Intelligence · doi:10.1007/s10489-024-06195-2 · Cited by 23
“It has been applied to a wide range of PDEs such as stochastic PDEs [6], fractional PDEs [7], Navier-Stokes [8].”
Wei, Y., Serra, Q., Gilles Lubineau, A., & Florentin, É. (2023) · Computers & Structures · doi:10.1016/j.compstruc.2023.107054 · Cited by 23
“Recently, PINNs have been applied to various types of PDEs [22–24].”
Cao, F., Gao, F., Guo, X., & Yuan, D. (2023) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2023.09.030 · Cited by 23
Chen, J., Song, J. M., Zhou, Z., & Yan, Z. (2023) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2023.114090 · Cited by 25
“Based on these advantages, the PINNs has also been applied extensively to different types of PDEs, including integro-differential equations [6], fractional PDEs [7], and stochastic PDEs [8, 9].”
MethodThe Deep Learning Galerkin Method for the General Stokes Equations
Li, J., Yue, J., Wen, Z., & Duan, W. (2022) · Journal of Scientific Computing · doi:10.1007/s10915-022-01930-8 · Cited by 23
“Besides, they proposed and developed a typical method physics-informed neural networks which combines observed data with PDE models [13, 14, 15] in many problems [16, 17, 18, 19, 20, 21].”
MethodPredicting the soliton dynamics and system parameters in optical fiber couplers
Yang, A., Xu, S.-Y., Liu, H., Li, N., & Sun, Y. (2024) · Nonlinear Dynamics · doi:10.1007/s11071-024-10235-4 · Cited by 22
“Lu et al. proposed a series of improvements to PINNs to optimize the neural network training process in terms of various aspects such as dataset, network form, physical constraints, etc. [28–30].”
Khadijeh, M., Cerqueglini, V., Kasbergen, C., Erkens, S. M. J. G., & Varveri, A. (2025) · Engineering With Computers · doi:10.1007/s00366-025-02174-4 · Cited by 22
“Corrosion modeling using PINNs successfully captured coupled elec - trochemistry and ion transport but struggled with moving boundary problems associated with corrosion front propagation [22 ].”
Lin, D. Y., & Yu, H.-Y. (2022) · Physical review. E · doi:10.1103/physreve.106.014503 · Cited by 23
SVD-PINNs: Transfer Learning of Physics-Informed Neural Networks via Singular Value Decomposition
Gao, Y., Cheung, K. C., & Ng, M. K. (2022) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci51031.2022.10022281 · Cited by 41
“Later, PINNs is extended to solve some more complicated and specific scientific computing problems, e.g., fractional PDEs [13], stochastic PDEs [2], [23], [25], [26] and high dimensional problems [20].”
BackgroundRiganti, R., & Negro, L. D. (2023) · Applied Physics Letters · doi:10.1063/5.0167155 · Cited by 26
Pu, H., Tan, B., Yi, J., Yuan, S., Zhao, J., Bai, R., & Luo, J. (2023) · Engineering With Computers · doi:10.1007/s00366-023-01914-8 · Cited by 22
“15, but we have some extra information on some points Γ i ⊂ Γ besides the governing equations and boundary conditions: From the implementation point of view, PINN solves inverse problems as easily as forward problems [30, 66].”
TNet: A Model-Constrained Tikhonov Network Approach for Inverse Problems
Nguyen, H. V., & Bui-Thanh, T. (2021) · SIAM Journal on Scientific Computing · doi:10.1137/22m1526708 · Cited by 19
Zou, J., Liu, C., Zhao, P., & Song, C. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3411472 · Cited by 22
“These include fractional PINN (fPINN) [11] for fractional equations, parareal PINN (PPINN) [12] designed to further enhance training efficiency for long-time physical problems, variational PINN (VPINN) [13] to reduce the order of differential operators, and PINN with hard constraints (hPINN) [14]…”
BackgroundSpeeding up and reducing memory usage for scientific machine learning via mixed precision
Hayford, J., Goldman-Wetzler, J., Wang, E., & Lu, L. (2024) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2024.117093 · Cited by 22
Self-citationMalki, I. E., Hakoume, A. E., Laghrib, A., & Hadri, A. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.129267 · Cited by 23
Metadata
- Authors
- Pang, G., Lu, L., & Karniadakis, G. E.
- Venue
- SIAM Journal on Scientific Computing
- Published
- 2019-01
- Volume
- 41
- Issue
- 4
- Pages
- A2603-A2626
- ISSN
- 1064-8275 (ISSN-L)
- 1095-7197
- OpenAlex
- W2900501930
- Cited by
- 1,105
- References
- 38
- Citations by year
- 2026: 253 · 2025: 260 · 2024: 195 · 2023: 155 · 2022: 119 · 2021: 82 · 2020: 35 · 2019: 6
- FWCI
- 28.01
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- SIAM Journal on Scientific Computing (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Army Research Office (W911NF-18-1-0301, W911NF-15-1-0562)
- U.S. Department of Energy (DE-SC0019434, DE-SC0019453)
- Topics
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Model Reduction and Neural Networks
- Physical Sciences › Mathematics › Modeling and Simulation › Fractional Differential Equations Solutions
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Keywords
- physics-informed neural networks, fractional advection-diffusion equation, fractional Poisson equation, convergence analysis, inverse problems, sampling error, optimization error
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.