pith. sign in

arxiv: 2606.18417 · v1 · pith:QPL7EA6Rnew · submitted 2026-06-16 · 💻 cs.CE

Enhancing neural network extrapolation in thermo-fluid systems using steady-state solutions

Sanjeeb Poudel , Teeratorn Kadeethum , Sanghyun Lee This is my paper

Pith reviewed 2026-06-26 21:37 UTC · model grok-4.3

classification 💻 cs.CE
keywords physics-informed neural networkstemporal extrapolationsteady-state solutionsdissipative PDEsthermo-fluid systemsNavier-Stokes equationsconjugate heat transfer
0
0 comments X

The pith

A neural network ansatz that decomposes solutions into a prescribed steady state plus a decaying transient correction embeds long-time convergence by construction for dissipative PDEs.

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

The paper introduces a representation for time-dependent PDE solutions in which the network output is written as the sum of a known steady-state field and a transient term multiplied by a decay function that goes to zero at large times. This structure guarantees that the solution approaches the steady state without requiring an extra loss term to enforce the asymptotics. The method is implemented inside a physics-informed neural network and tested on problems that range from the heat equation to lid-driven cavity flow, natural convection, and three-dimensional conjugate heat transfer. Results indicate that the built-in convergence improves accuracy when the network is asked to predict times beyond the interval used for training. A reader would care because standard networks commonly lose fidelity on long-time forecasts, and the new form addresses that limitation directly through the functional form rather than through added penalties.

Core claim

The proposed ansatz decomposes the solution into a steady-state component and a transient correction modulated by a time-dependent decay profile. When the decay profile vanishes at long time and the transient correction remains bounded, the representation embeds convergence to the prescribed steady state directly into the architecture rather than enforcing it through an additional penalty term. This allows the network to learn the transient dynamics while preserving the correct asymptotic behavior.

What carries the argument

The steady-state-informed ansatz that writes the solution as the sum of a prescribed steady-state field and a bounded transient correction multiplied by a decay profile that vanishes at large times.

If this is right

  • The network only needs to learn the transient correction; the long-time limit is satisfied automatically by the functional form.
  • Temporal extrapolation beyond the training interval improves compared with networks that lack the explicit asymptotic embedding.
  • The approach applies to dissipative systems ranging from the one-dimensional heat equation to incompressible Navier-Stokes flow and three-dimensional conjugate heat transfer.
  • Training uses the physics-informed loss inside a PINN framework without an additional steady-state penalty.

Where Pith is reading between the lines

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

  • The same decomposition could be tried on systems whose steady states are computed once by a separate solver and then frozen for the network.
  • If the decay profile itself is learned rather than prescribed, the method might adapt to cases where the relaxation rate is not known in advance.
  • The construction may reduce the amount of long-time data needed during training because the network is not free to drift away from the known equilibrium.

Load-bearing premise

The PDE solutions relax toward a stationary equilibrium that can be prescribed or computed independently and supplied to the network.

What would settle it

A numerical example in which the learned transient correction grows unbounded even though the decay profile vanishes would show that the architectural guarantee of convergence does not hold.

Figures

Figures reproduced from arXiv: 2606.18417 by Sanghyun Lee, Sanjeeb Poudel, Teeratorn Kadeethum.

Figure 1
Figure 1. Figure 1: Proposed network architecture and multi-phase training strategy. Phase 1 (Steady-State): Fully connected [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: One-dimensional heat equation: Extrapolation performance and temporal function [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: One-dimensional heat equation: Extrapolated solution versus the reference solution for [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: One-dimensional heat equation: Extrapolation error as a function of training duration. The vertical axis shows [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Lid-Driven Cavity Flow: L∞ error over time for the velocity components (a) u1 and (b) u2, evaluated across various temporal functions f(t), with the network trained on t ∈ [0, 0.5]. 4.2.1 Training on t ∈ [0, 0.5] [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Lid-Driven Cavity Flow: L∞ error over time for the velocity components (a) u1 and (b) u2, evaluated across various temporal functions f(t), with the network trained on t ∈ [0, 1]. times the training horizon (t = 1.5 versus training up to t = 0.5), demonstrates that the steady-state-informed ansatz provides meaningful predictive capability well outside the training domain. (a) u h 1 (b) |u1 − u h 1 | (c) u … view at source ↗
Figure 7
Figure 7. Figure 7: Lid-Driven Cavity Flow: Velocity predictions and pointwise absolute errors at [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Lid-Driven Cavity Flow: Contour plots showing the absolute spatial differences in the velocity components [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Natural convection in a square cavity: (a) [PITH_FULL_IMAGE:figures/full_fig_p014_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Natural convection in a square cavity: Temporal evolution of temperature profiles (top row) and the [PITH_FULL_IMAGE:figures/full_fig_p015_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Three-dimensional domain for conjugate heat transfer [PITH_FULL_IMAGE:figures/full_fig_p017_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Conjugate heat transfer in three dimensions: (a) Relative [PITH_FULL_IMAGE:figures/full_fig_p018_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Conjugate heat transfer in three dimensions: (a) Relative [PITH_FULL_IMAGE:figures/full_fig_p018_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Conjugate heat transfer in three dimensions: Temporal evolution of temperature profiles (top row) on a YZ [PITH_FULL_IMAGE:figures/full_fig_p019_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Conjugate heat transfer in three dimensions: The top row illustrates the temporal evolution of temperature [PITH_FULL_IMAGE:figures/full_fig_p020_15.png] view at source ↗
read the original abstract

Time-dependent partial differential equations (PDEs) arise in many engineering systems, including thermo-fluid applications. Classical numerical simulations of such systems can become computationally expensive for long-time dynamics because they typically require sequential time integration with time steps constrained by stability, accuracy, or nonlinear solvers. Although scientific machine learning provides an alternative for approximating PDE solutions, standard neural network approximations often degrade when extrapolated beyond the training time interval. In this work, we propose a steady-state-informed neural network representation for dissipative PDE systems whose solutions relax toward a stationary equilibrium. The proposed ansatz decomposes the solution into a steady-state component and a transient correction modulated by a time-dependent decay profile. When the decay profile vanishes at long time and the transient correction remains bounded, the representation embeds convergence to the prescribed steady state directly into the architecture, rather than enforcing it through an additional penalty term. This allows the network to learn the transient dynamics while preserving the correct asymptotic behavior. We implement the approach within a physics-informed neural network (PINN) framework and train the resulting model using the SOAP optimizer. The method is evaluated on a sequence of problems of increasing physical and geometric complexity, ranging from the one-dimensional heat equation to incompressible Navier-Stokes flow in a lid-driven cavity, natural convection in a square cavity, and a full three-dimensional conjugate heat transfer problem. The numerical results show that the steady-state-informed architecture substantially improves temporal extrapolation beyond the training interval compared with architectures that do not explicitly enforce the asymptotic condition.

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

1 major / 0 minor

Summary. The paper proposes a steady-state-informed neural network ansatz for dissipative time-dependent PDEs whose solutions relax to a stationary equilibrium. The solution is decomposed as a prescribed steady-state component plus a neural-network-modeled transient correction modulated by a decay profile that vanishes at long times. When the transient remains bounded, this embeds asymptotic convergence directly into the architecture rather than via an added penalty. The ansatz is implemented in a PINN framework trained with the SOAP optimizer and evaluated on a sequence of problems of increasing complexity (1D heat equation, lid-driven cavity, natural convection, 3D conjugate heat transfer), with the claim that it substantially improves temporal extrapolation beyond the training interval relative to architectures without the explicit asymptotic constraint.

Significance. If the boundedness assumption holds and is validated, the approach offers a non-circular architectural mechanism for enforcing correct long-time behavior in neural approximations of relaxing thermo-fluid systems. This could reduce the need for penalty terms and improve reliability of extrapolation in engineering contexts where classical time-stepping is expensive. The explicit separation of steady state from transient distinguishes the method from standard PINNs and is a strength when the premise applies.

major comments (1)
  1. [Abstract] Abstract: the central claim that the representation 'embeds convergence to the prescribed steady state directly into the architecture' (rather than through a penalty) requires that the NN-modeled transient correction remain bounded for t > T. The transient network is trained exclusively on a finite interval [0, T]; no architectural constraint, regularization, or post-training verification is described that would enforce or confirm boundedness outside the training window. If the transient grows (even slowly) for t > T, the embedded convergence property fails and the reported extrapolation benefit is lost. This assumption is load-bearing for the distinction from penalty-based methods and must be addressed with either a proof, a boundedness-enforcing modification, or explicit numerical verification on the test cases.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed and constructive feedback, which highlights a critical assumption in our proposed ansatz. We address the major comment point-by-point below and commit to revisions that strengthen the manuscript without altering its core claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the representation 'embeds convergence to the prescribed steady state directly into the architecture' (rather than through a penalty) requires that the NN-modeled transient correction remain bounded for t > T. The transient network is trained exclusively on a finite interval [0, T]; no architectural constraint, regularization, or post-training verification is described that would enforce or confirm boundedness outside the training window. If the transient grows (even slowly) for t > T, the embedded convergence property fails and the reported extrapolation benefit is lost. This assumption is load-bearing for the distinction from penalty-based methods and must be addressed with either a proof, a boundedness-enforcing modification, or explicit numerical verification on the test cases.

    Authors: We agree that boundedness of the transient correction for t > T is a load-bearing assumption for the architectural (rather than penalty-based) enforcement of asymptotic behavior, and that the manuscript states this condition explicitly but does not provide supporting verification. A general proof of boundedness is not available without further restrictions on the network architecture or activation functions, and we do not introduce an additional constraint or regularization term, as this would change the method. In the revised version we will add explicit numerical verification: for each test case we will plot or tabulate the L2 norm of the transient network output over an extended extrapolation interval (e.g., up to 2T or 3T) to confirm that it remains bounded, thereby substantiating the claim for the dissipative systems considered. This verification will be included in the results section and discussed in the abstract revision. revision: yes

Circularity Check

0 steps flagged

No significant circularity; ansatz is an explicit architectural choice with external inputs.

full rationale

The paper's central step is the introduction of an explicit decomposition ansatz (solution = steady-state component + decay-modulated transient correction) whose convergence property holds by the algebraic form of the ansatz when the decay vanishes and the transient term is bounded. The steady-state is supplied externally or computed independently, not fitted from the target transient data. The transient correction is represented by a standard neural network trained on the usual residual loss; no claimed prediction is obtained by renaming or refitting a quantity already present in the inputs. No self-citation chain, uniqueness theorem, or smuggled ansatz is invoked to justify the form. Numerical results on multiple PDEs therefore constitute independent validation rather than a tautological restatement of fitted quantities.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that the target PDE systems are dissipative and converge to a known stationary state, plus the modeling choice of a bounded transient correction modulated by a decay profile whose functional form is not detailed in the abstract.

free parameters (1)
  • decay profile parameters
    The time-dependent decay profile is introduced as part of the ansatz; its specific functional form and any tunable coefficients are not specified in the abstract.
axioms (1)
  • domain assumption Solutions of the considered dissipative PDE systems relax toward a stationary equilibrium
    Invoked in the abstract to delimit the class of problems for which the ansatz is designed.

pith-pipeline@v0.9.1-grok · 5804 in / 1436 out tokens · 33645 ms · 2026-06-26T21:37:54.285426+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

54 extracted references · 5 canonical work pages · 3 internal anchors

  1. [1]

    SIAM, 2007

    Randall J LeVeque.Finite difference methods for ordinary and partial differential equations: steady-state and time-dependent problems. SIAM, 2007

    2007

  2. [2]

    LeVeque.Finite Volume Methods for Hyperbolic Problems

    Randall J. LeVeque.Finite Volume Methods for Hyperbolic Problems. Cambridge University Press, 2002

    2002

  3. [3]

    Pearson Education, 2nd edition, 2007

    Henk Kaarle Versteeg and Weeratunge Malalasekera.An Introduction to Computational Fluid Dynamics: The Finite Volume Method. Pearson Education, 2nd edition, 2007

    2007

  4. [4]

    Thomas J. R. Hughes.The Finite Element Method: Linear Static and Dynamic Finite Element Analysis. Dover Publications, 2012. Originally published by Prentice-Hall, 1987. 20

    2012

  5. [5]

    Zienkiewicz, Robert L

    Olek C. Zienkiewicz, Robert L. Taylor, and Jian Z. Zhu.The Finite Element Method: Its Basis and Fundamentals. Elsevier Butterworth-Heinemann, 6th edition, 2005

    2005

  6. [6]

    Springer, 2nd edition, 2009

    Alfio Quarteroni, Riccardo Sacco, and Fausto Saleri.Numerical Mathematics. Springer, 2nd edition, 2009

    2009

  7. [7]

    J. N. Reddy.Introduction to the Finite Element Method. McGraw-Hill Education, 4th edition, 2019

    2019

  8. [8]

    Dynamic thermal modelling of power transformers.IEEE Transactions on Power Delivery, 20(1):197–204, 2005

    Dejan Susa, Matti Lehtonen, and Hasse Nordman. Dynamic thermal modelling of power transformers.IEEE Transactions on Power Delivery, 20(1):197–204, 2005

    2005

  9. [9]

    Seddik, Jehan Shazly, and Magdy B

    Mohamed S. Seddik, Jehan Shazly, and Magdy B. Eteiba. Thermal analysis of power transformer using 2D and 3D finite element method.Energies, 17(13):3203, 2024

    2024

  10. [10]

    Numerical investigation of oil flow and temperature distributions for ON transformer windings.Applied Thermal Engineering, 130:1–9, 2018

    Xiang Zhang, Zhongdong Wang, and Qiang Liu. Numerical investigation of oil flow and temperature distributions for ON transformer windings.Applied Thermal Engineering, 130:1–9, 2018

    2018

  11. [11]

    Enhancing heat transfer in power transformer radiators via thermo-fluid dynamic analysis with periodic thermal boundary conditions

    Jonathan J Dorella, Bruno A Storti, Gustavo A Ríos Rodriguez, and Mario A Storti. Enhancing heat transfer in power transformer radiators via thermo-fluid dynamic analysis with periodic thermal boundary conditions. International Journal of Heat and Mass Transfer, 222:125142, 2024

    2024

  12. [12]

    Impacts of startup, shutdown and load variation on transient temperature and thermal stress fields within blades of gas turbines.Journal of Thermal Science, 31(3):727–740, 2022

    Liuxi Cai, Yanfang Hou, Fang Li, Yun Li, Shunsen Wang, and Jingru Mao. Impacts of startup, shutdown and load variation on transient temperature and thermal stress fields within blades of gas turbines.Journal of Thermal Science, 31(3):727–740, 2022

    2022

  13. [13]

    W. Z. Tang, L. Yang, W. Zhu, et al. Numerical simulation of temperature distribution and thermal-stress field in a turbine blade with multilayer-structure TBCs by a fluid-solid coupling method.Journal of Materials Science & Technology, 32(5):452–458, 2016

    2016

  14. [14]

    High-re solutions for incompressible flow using the navier-stokes equations and a multigrid method.Journal of computational physics, 48(3):387–411, 1982

    UKNG Ghia, Kirti N Ghia, and CT Shin. High-re solutions for incompressible flow using the navier-stokes equations and a multigrid method.Journal of computational physics, 48(3):387–411, 1982

    1982

  15. [15]

    P. N. Shankar and M. D. Deshpande. Fluid mechanics in the driven cavity.Annual Review of Fluid Mechanics, 32(1):93–136, 2000

    2000

  16. [16]

    Wiley New York, 1996

    Frank P Incropera, David P DeWitt, Theodore L Bergman, Adrienne S Lavine, et al.Fundamentals of heat and mass transfer, volume 6. Wiley New York, 1996

    1996

  17. [17]

    Recent advances in non-intrusive polynomial chaos and stochastic collocation methods for uncertainty analysis and design

    Michael Eldred. Recent advances in non-intrusive polynomial chaos and stochastic collocation methods for uncertainty analysis and design. In50th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference 17th AIAA/ASME/AHS Adaptive Structures Conference 11th AIAA No, page 2274, 2009

    2009

  18. [18]

    Princeton university press, 2010

    Dongbin Xiu.Numerical methods for stochastic computations: a spectral method approach. Princeton university press, 2010

    2010

  19. [19]

    Physics- informed machine learning.Nature Reviews Physics, 3(6):422–440, 2021

    George Em Karniadakis, Ioannis G Kevrekidis, Lu Lu, Paris Perdikaris, Sifan Wang, and Liu Yang. Physics- informed machine learning.Nature Reviews Physics, 3(6):422–440, 2021

    2021

  20. [20]

    Brunton and J

    Steven L. Brunton and J. Nathan Kutz.Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control. Cambridge University Press, 2nd edition, 2022

    2022

  21. [21]

    Karniadakis

    Maziar Raissi, Paris Perdikaris, and George E. Karniadakis. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations.Journal of Computational Physics, 378:686–707, 2019

    2019

  22. [22]

    Jørgensen, and Hamidreza M

    Teeratorn Kadeethum, Thomas M. Jørgensen, and Hamidreza M. Nick. Physics-informed neural networks for solving nonlinear diffusivity and Biot’s equations.PloS one, 15(5):e0232683, 2020

    2020

  23. [23]

    Jagtap, Ehsan Kharazmi, and George Em Karniadakis

    Ameya D. Jagtap, Ehsan Kharazmi, and George Em Karniadakis. Conservative physics-informed neural networks on discrete domains for conservation laws: Applications to forward and inverse problems.Computer Methods in Applied Mechanics and Engineering, 365:113028, 2020

    2020

  24. [24]

    Jørgensen, and Hamidreza M

    Teeratorn Kadeethum, Thomas M. Jørgensen, and Hamidreza M. Nick. Physics-informed neural networks for solv- ing inverse problems of nonlinear Biot’s equations: batch training. InARMA US Rock Mechanics/Geomechanics Symposium, pages ARMA–2020. ARMA, 2020

    2020

  25. [25]

    DeepXDE: A deep learning library for solving differential equations.SIAM Review, 63(1):208–228, 2021

    Lu Lu, Xuhui Meng, Zhiping Mao, and George Em Karniadakis. DeepXDE: A deep learning library for solving differential equations.SIAM Review, 63(1):208–228, 2021

    2021

  26. [26]

    Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators.Nature Machine Intelligence, 3(3):218–229, 2021

    Lu Lu, Pengzhan Jin, Guofei Pang, Zhongqiang Zhang, and George Em Karniadakis. Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators.Nature Machine Intelligence, 3(3):218–229, 2021. 21

    2021

  27. [27]

    Fourier neural operator for parametric partial differential equations

    Zongyi Li, Nikolas Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhatt, Andrew Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differential equations. InInternational Conference on Learning Representations, 2021

    2021

  28. [28]

    Learning the solution operator of parametric partial differential equations with physics-informed DeepONets.Science Advances, 7(40):eabi8605, 2021

    Sifan Wang, Hanwen Wang, and Paris Perdikaris. Learning the solution operator of parametric partial differential equations with physics-informed DeepONets.Science Advances, 7(40):eabi8605, 2021

    2021

  29. [29]

    Neural operator: Learning maps between function spaces with applications to PDEs.Journal of Machine Learning Research, 24(89):1–97, 2023

    Nikolas Kovachki, Zongyi Li, Burigede Liu, Kamyar Azizzadenesheli, Kaushik Bhatt, Andrew Stuart, and Anima Anandkumar. Neural operator: Learning maps between function spaces with applications to PDEs.Journal of Machine Learning Research, 24(89):1–97, 2023

    2023

  30. [30]

    When and why PINNs fail to train: A neural tangent kernel perspective.Journal of Computational Physics, 449:110768, 2022

    Sifan Wang, Xinling Yu, and Paris Perdikaris. When and why PINNs fail to train: A neural tangent kernel perspective.Journal of Computational Physics, 449:110768, 2022

    2022

  31. [31]

    Aditi Krishnapriyan, Amir Gholami, Shandian Zhe, Robert Kirby, and Michael W. Mahoney. Characterizing possible failure modes in physics-informed neural networks. 34:26548–26560, 2021

    2021

  32. [32]

    Revanth Mattey and Susanta Ghosh. A novel sequential method to train physics informed neural networks for Allen–Cahn and Cahn–Hilliard equations.Computer Methods in Applied Mechanics and Engineering, 390:114474, 2022

    2022

  33. [33]

    Understanding and mitigating gradient flow pathologies in physics-informed neural networks.SIAM Journal on Scientific Computing, 43(5):A3055–A3081, 2021

    Sifan Wang, Yujun Teng, and Paris Perdikaris. Understanding and mitigating gradient flow pathologies in physics-informed neural networks.SIAM Journal on Scientific Computing, 43(5):A3055–A3081, 2021

    2021

  34. [34]

    Chenxi Wu, Min Zhu, Qinyang Tan, Yadhu Kartha, and Lu Lu. A comprehensive study of non-adaptive and residual-based adaptive sampling for physics-informed neural networks.Computer Methods in Applied Mechanics and Engineering, 403:115671, 2023

    2023

  35. [35]

    SOAP: Improving and Stabilizing Shampoo using Adam

    Nikhil Vyas, Depen Morwani, Rosie Zhao, Mujin Kwun, Itai Shapira, David Brandfonbrener, Lucas Janson, and Sham Kakade. Soap: Improving and stabilizing shampoo using adam.arXiv preprint arXiv:2409.11321, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  36. [36]

    Gradient alignment in physics-informed neural networks: A second-order optimization perspective.arXiv preprint arXiv:2502.00604, 2025

    Sifan Wang, Ananyae Kumar Bhartari, Bowen Li, and Paris Perdikaris. Gradient alignment in physics-informed neural networks: A second-order optimization perspective.arXiv preprint arXiv:2502.00604, 2025

    work page arXiv 2025

  37. [37]

    MIT press Cambridge, 2016

    Ian Goodfellow, Yoshua Bengio, Aaron Courville, and Yoshua Bengio.Deep learning, volume 1. MIT press Cambridge, 2016

    2016

  38. [38]

    Searching for Activation Functions

    Prajit Ramachandran, Barret Zoph, and Quoc V Le. Searching for activation functions.arXiv preprint arXiv:1710.05941, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  39. [39]

    Multilayer feedforward networks are universal approxi- mators.Neural Networks, 2(5):359–366, 1989

    Kurt Hornik, Maxwell Stinchcombe, and Halbert White. Multilayer feedforward networks are universal approxi- mators.Neural Networks, 2(5):359–366, 1989

    1989

  40. [40]

    Approximation by superpositions of a sigmoidal function.Mathematics of Control, Signals and Systems, 2(4):303–314, 1989

    George Cybenko. Approximation by superpositions of a sigmoidal function.Mathematics of Control, Signals and Systems, 2(4):303–314, 1989

    1989

  41. [41]

    The expressive power of neural networks: A view from the width.Advances in neural information processing systems, 30, 2017

    Zhou Lu, Hongming Pu, Feicheng Wang, Zhiqiang Hu, and Liwei Wang. The expressive power of neural networks: A view from the width.Advances in neural information processing systems, 30, 2017

    2017

  42. [42]

    Pearlmutter, Alexey Andreyevich Radul, and Jeffrey Mark Siskind

    Atilim Gunes Baydin, Barak A. Pearlmutter, Alexey Andreyevich Radul, and Jeffrey Mark Siskind. Automatic differentiation in machine learning: a survey.Journal of Machine Learning Research, 18(153):1–43, 2018

    2018

  43. [43]

    Respecting causality for training physics-informed neural networks.Computer Methods in Applied Mechanics and Engineering, 421:116813, 2024

    Sifan Wang, Shyam Sankaran, and Paris Perdikaris. Respecting causality for training physics-informed neural networks.Computer Methods in Applied Mechanics and Engineering, 421:116813, 2024

    2024

  44. [44]

    Adam: A Method for Stochastic Optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  45. [45]

    Shampoo: Preconditioned stochastic tensor optimization

    Vineet Gupta, Tomer Koren, and Yoram Singer. Shampoo: Preconditioned stochastic tensor optimization. In International Conference on Machine Learning, pages 1842–1850. PMLR, 2018

    2018

  46. [46]

    Adafactor: Adaptive learning rates with sublinear memory cost

    Noam Shazeer and Mitchell Stern. Adafactor: Adaptive learning rates with sublinear memory cost. InInternational conference on machine learning, pages 4596–4604. PMLR, 2018

    2018

  47. [47]

    Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, Stéfan J

    Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, Stéfan J. van der Walt, Matthew Brett, Joshua Wilson, K. Jarrod Millman, Nikolay Mayorov, Andrew R. J. Nelson, Eric Jones, Robert Kern, Eric Larson, C J Carey, ˙Ilhan Polat, Yu Feng, Eric W. M...

    2020

  48. [48]

    Harris, K

    Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J. Smith, Robert Kern, Matti Picus, Stephan Hoyer, Marten H. van Kerkwijk, Matthew Brett, Allan Haldane, Jaime Fernández del Río, Mark Wiebe, Pearu Peterson, Pierre Gérard-Marchant, Kevin Shepp...

    2020

  49. [49]

    Keras.https://keras.io, 2015

    François Chollet et al. Keras.https://keras.io, 2015

    2015

  50. [50]

    Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S. Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Ian Goodfellow, Andrew Harp, Geoffrey Irving, Michael Isard, Yangqing Jia, Rafal Jozefowicz, Lukasz Kaiser, Manjunath Kudlur, Josh Levenberg, Dandelion Mané, Rajat Monga, Sherry Moore, Derek Murra...

    2015

  51. [51]

    J. D. Hunter. Matplotlib: A 2d graphics environment.Computing in Science & Engineering, 9(3):90–95, 2007

    2007

  52. [52]

    Pressure-robust enriched galerkin finite element methods for coupled navier-stokes and heat equations.arXiv preprint arXiv:2512.16716, 2025

    Sanjeeb Poudel, Sanghyun Lee, and Lin Mu. Pressure-robust enriched galerkin finite element methods for coupled navier-stokes and heat equations.arXiv preprint arXiv:2512.16716, 2025

    work page arXiv 2025

  53. [53]

    The deal

    Daniel Arndt, Wolfgang Bangerth, Maximilian Bergbauer, Marco Feder, Marc Fehling, Johannes Heinz, Timo Heister, Luca Heltai, Martin Kronbichler, Matthias Maier, et al. The deal. ii library, version 9.5.Journal of Numerical Mathematics, 31(3):231–246, 2023

    2023

  54. [54]

    Gauthier-Villars, 1897

    Joseph Boussinesq.Théorie de l’écoulement tourbillonnant et tumultueux des liquides dans les lits rectilignes à grande section..., volume 1. Gauthier-Villars, 1897. 23