betaFeedback

WorksMathematics

Variationally correct physics-informed operator learning for nonlinear partial differential equations

Abstract

Physics-informed operator learning trains neural operators on the residual of a parametric partial differential equation (PDE) instead of labeled solution data, but its reliability depends on a variationally correct residual: one that is equivalent to the error in a norm dictated by the equation, the PDE-compliant norm. For nonlinear equations written as first-order systems, we prove that the first-order system least-squares residual is equivalent to the error in this norm near a solution branch if the linearized residual is stable and the residual derivative is Lipschitz continuous, possibly in a norm stronger than the error norm. We verify these conditions for monotone reaction, solution- and gradient-dependent diffusion, and compressible Neo-Hookean hyperelasticity. The stability mechanisms vary across these classes, as do the regions where the equivalence holds, from arbitrary bounded sets to a neighborhood of the branch. In a conforming finite element space, we show that the loss excess of a state near the finite element solution is equivalent to its squared distance from that solution. We use this loss to train a reduced basis neural operator with values in this space, with solution snapshots needed only for the basis and the loss and its symbolically derived gradient assembled by the finite element library. At inference, this loss estimates the error. Numerically, the square root of the loss excess tracks the error within factors comparable to those of the linear case, and the operator achieves higher accuracy in the PDE-compliant norm than grid-based baselines evaluated in the same finite element space.

Introduction

Uncertainty quantification, inverse problems, and design optimization require the solution of a parametric partial differential equation (PDE) for many parameter values. For nonlinear PDEs, each solve typically requires an iterative procedure, making solution evaluation across many parameter values computationally expensive. Neural operators such as the deep operator network (DeepONet) [1] and the Fourier neural operator (FNO) [2] provide fast surrogates for these solves by learning the parameter-to-solution map. Data-driven neural operators learn from solution snapshots, shifting the computational cost to an offline training stage that may itself require many PDE solves. Their computational advantage over classical solvers depends on the problem structure and target accuracy, with neural operators being most competitive at low-to-moderate accuracy requirements, even when data-generation and training costs are fully amortized [3]. Alternatively, physics-informed operator learning trains on the PDE residual and avoids the need for solution snapshots, since the residual of a prediction can be evaluated at any parameter value. It extends physics-informed neural networks [4] and their variational formulations [5] from a single PDE instance to the parameter-to-solution map [6, 7]. The usefulness of this approach hinges on two questions: Does a small residual imply a small error, and in which norm? How can the residual be evaluated and differentiated accurately and efficiently enough to serve as a training loss?

For linear problems, the answer to the first question rests on a stable variational formulation, in which errors in the trial norm are equivalent to residuals in the dual test norm. For several PDE models, a first-order formulation is stable with a test space of L2L^2 type, which is self-dual. If H\mathsf{H} is the state space, s∗(p)s^*(\mathrm{p}) the solution for the parameter p\mathrm{p}, and Bps−Sp\mathcal{B}_{\mathrm{p}}s-S_{\mathrm{p}} the residual in an L2L^2-type product space L2L^2, then

c0∥s−s∗(p)∥H2≤∥Bps−Sp∥L22≤C0∥s−s∗(p)∥H2,(1)c_0\lVert s-s^*(\mathrm{p})\rVert_{\mathsf{H}}^2\leq\lVert\mathcal{B}_{\mathrm{p}}s-S_{\mathrm{p}}\rVert_{L^2}^2\leq C_0\lVert s-s^*(\mathrm{p})\rVert_{\mathsf{H}}^2, \tag*{(1)}

so that the first-order system least-squares (FOSLS) residual loss is equivalent to the error in a norm dictated by the PDE [8, 9, 10, 11]. Losses with this property are called variationally correct [12, 13]. In [14] we combined a FOSLS loss, variational lifts of mixed boundary data, and a reduced basis neural operator (RBNO) that predicts reduced basis coefficients in a conforming finite element space for the full first-order state. Fluxes and stresses are direct outputs of the model rather than quantities recovered by postprocessing the predicted primal field. Each prediction can be evaluated with the same residual functional as the finite element solution, and solution snapshots are needed only to construct the reduced basis. In contrast to data-driven reduced models [15, 16, 17], the network is trained by minimizing the residual loss, which provides an estimate of the prediction error.

The choice of residual decides whether a physics-informed loss controls the error. A pointwise strong-form residual can fail to distinguish the intended PDE solution from solutions of a different problem, allowing a neural network to achieve a small training loss while approximating the wrong solution. In the first benchmark of § 6, we consider −div⁡(p∇u)+κu3=f-\operatorname{div}(p\nabla u)+\kappa u^3=f, with f=f2+div⁡f1f=f_2+\operatorname{div}f_1, where the conductivity pp and the field f1f_1 are piecewise constant and jump across the boundaries of sixteen square inclusions. Away from these interfaces, their derivatives vanish, so pointwise evaluation of the second-order residual, as commonly used in physics-informed neural networks, yields only −pΔu+κu3−f2-p\Delta u+\kappa u^3-f_2. This residual does not capture the interface contribution of f1f_1 or enforce the required transmission condition. It therefore vanishes not only at the intended solution, but also at the solution of the problem with f1f_1 omitted, and more generally at any sufficiently regular function satisfying the equation within each subdomain and the external boundary conditions, without necessarily satisfying the transmission condition. The first-order formulation addresses this ambiguity by requiring the total flux σ=p∇u+f1\sigma=p\nabla u+f_1 to belong to H(div⁡;Ω)H(\operatorname{div};\Omega), which ensures continuity of its normal component across the interfaces, as discussed in Remark 2.1. Moreover, Proposition 4.1 shows that the resulting FOSLS residual controls the H1H^1 error of the primal component uu.

In this paper, we extend the variationally correct physics-informed operator learning framework developed in [14] from linear to nonlinear PDEs. This extension requires addressing two main issues. First, a residual–error equivalence such as (1) generally holds only locally. Nonlinear equations may admit multiple solutions, their linearizations may become singular along a solution branch, and constitutive laws may restrict the admissible states. Second, establishing this equivalence requires assumptions that depend on the form of the nonlinearity. For reaction terms, which enter the lower-order part of the operator, monotonicity provides the needed control. For solution- and gradient-dependent diffusion, the nonlinearity enters the principal part, and Lipschitz continuity of the linearization generally requires a stronger norm than the one used to measure the error. In finite-strain elasticity, the stress law is nonpolynomial and defined only for orientation-preserving deformations, while the linearized operator may lose stability as the load varies. To accommodate these different structures, we express the residual as the sum of a linear first-order operator and a nonlinear term acting on the primal variable. We then identify two conditions on the linearization that ensure local residual–error equivalence and verify these conditions for each of the three problem classes.

Local stability analysis has its roots in the approximation theory of nonsingular solution branches [18], and residual-based error estimates are well established for nonlinear reduced basis methods [19, 20, 21]. In our setting, the loss measures the residual of the full first-order state and serves as the training objective for a parameter-to-solution map. The residual–error equivalence must therefore hold uniformly over the parameter set and remain valid under reduced basis approximation and numerical integration. Related questions for individual PDE instances have been studied through neural least-squares methods [22, 23], nonlinear least-squares finite element methods [24], and a posteriori error bounds for physics-informed neural networks [25]. For operator learning, recent work has connected residual-based adaptivity to variational objectives [26] and trained reduced basis neural operators using physics-informed residuals [27].

The second question concerns how to evaluate and differentiate the residual efficiently during training. In the linear case, the loss of a reduced state is a quadratic form with matrices that can be assembled once [14]. For nonlinear problems, this structure is generally unavailable, so we retain finite element assembly in the training loop. For each network prediction, the finite element code evaluates the loss and assembles its gradient with respect to the finite element coefficients (or degrees of freedom). The corresponding derivative form is obtained by symbolic differentiation and compiled once by the finite element library. This gradient is then propagated to the network parameters through the chain rule [28, 29], without requiring a forward or adjoint solve. Galerkin residuals have previously been used as data-free training losses for operator networks [30], with an error analysis for linear elliptic problems involving the condition number of the finite element matrix [31]. In our formulation, the loss is the squared L2L^2 norm of the first-order residual evaluated at a conforming state. Under the local stability assumptions, it is equivalent to the squared error in the PDE-compliant norm, with constants independent of the mesh.

Contributions

The main contributions are as follows:

  1. Variational correctness for nonlinear first-order systems. In Theorem 3.1, we establish local equivalence between the FOSLS residual and the solution error under stability and Lipschitz continuity assumptions on the linearization. The error is measured in the PDE-compliant norm of the first-order system, while the neighborhood of the solution branch may be defined in a stronger norm to accommodate gradient-dependent nonlinearities. We further account for errors in residual evaluation through an additive term in Proposition 3.2. Within this neighborhood, the residual therefore provides an assessment of prediction accuracy without requiring a reference solution.

  1. Verification for three classes of nonlinearity. We verify these assumptions for monotone reaction, nonlinear diffusion with solution-dependent coefficients of Kirchhoff type or gradient-dependent coefficients of pp-Laplace type, and compressible Neo-Hookean hyperelasticity, with the results summarized in Table 2. For monotone reaction and Kirchhoff-type diffusion, residual–error equivalence holds on every bounded set of states. For the reaction problem, we further obtain an error bound computable from the problem data and the prediction. For the hyperelasticity problem, the equivalence holds locally near the solution branch and relies on the stability of the tangent operator, which we assess numerically.

Semilinear diffusion–reaction (Section 4.1)Quasilinear diffusion (Section 4.2)Compressible hyperelasticity (Section 4.3)
Nonlinear termreaction κu3\kappa u^3flux p(1+u2)∇u\mathrm{p}(1+u^2)\nabla ustress P(I+∇u;p)P(I+\nabla u;\mathbf{p})
It entersthe lower-order termthe principal partthe principal part; defined only for det⁡F>0\det F>0
Norm of the neighborhood Np\mathcal{N}_{\mathbf{p}} (Assumption 2)H1H^1, the norm of the errorW1,∞W^{1,\infty}W1,∞W^{1,\infty}, with det⁡F≥j0\det F\ge j_0
Stability of the linearization (Assumption 1)proved: monotonicity (Proposition 4.1)proved: factorization of the tangent (Proposition 4.4)assumed: coercive or invertible tangent at the branch (Proposition 4.5)
Estimate holds onevery bounded set in H1H^1every bounded set in W1,∞W^{1,\infty}a neighborhood of the branch
Computable error boundyes (Corollary 4.2)given a bound M∗M_* on the solution in W1,∞W^{1,\infty}no; a∗a_* and C∗C_* are not known
Benchmark (Table 3)heat conduction with a cubic reactionflow with a solution-dependent permeabilityNeo-Hookean body under traction

Table 2. The three classes side by side: how the nonlinear term enters, the two properties required by Theorem 3.1, and what they imply for the residual–error estimate. Each column corresponds to one subsection and to one benchmark of Section 6.

  1. Physics-informed training through finite element residuals. In Lemma 5.1, we show that, for conforming finite element states near the finite element solution, the loss excess relative to that solution is equivalent to the squared distance from it, allowing the equivalence constants to be assessed numerically. We use this loss to train RBNO through finite element assembly and differentiation, without nonlinear solves during training or solution data beyond the snapshots used to construct the reduced basis. The same procedure applies to other predictors whose outputs lie in the conforming space. In our benchmarks, one evaluation of the loss and its gradient costs between 1/500 and 1/30 of a finite element solve.

  1. Numerical validation across the three problem classes. We study semilinear diffusion–reaction, quasilinear diffusion, and compressible hyperelasticity benchmarks using the same data and discretizations as the corresponding linear benchmarks in [14]. The solution error and the square root of the loss excess remain comparable, with ratios similar to those observed in the linear case. Discrete stability tests also support the stability assumptions at the computed states. When evaluated in the same finite element space, RBNO achieves smaller PDE-compliant norm errors than grid-based baselines, including PCA-Net, FNO, and FC-PINO. These comparisons further show that small grid-based residuals can coexist with substantial PDE-compliant norm errors, even when the predictions are accurate in L2L^2.

Organization

The remainder of the paper is organized as follows. We introduce the residual formulation with boundary lifting in § 2. In § 3, we establish local residual–error equivalence and account for errors in residual evaluation. We then verify the assumptions for three classes of nonlinear PDEs in § 4, with the main results summarized in Table 2. In § 5, we develop the conforming finite element formulation, derive loss-excess estimates, and describe the discrete stability assessment and RBNO training procedure. We present numerical experiments in § 6 to validate the error estimates, examine stability at the computed states, and compare RBNO with grid-based neural operators. Finally, we conclude in § 7. Readers primarily interested in the method and numerical results may focus on § 2, the statement of Theorem 3.1, Table 2, and § 5 and § 6.

Parametric nonlinear FOSLS formulation

We formulate each model as a first-order system for the flux or stress σ\sigma and the primal variable uu. We remove the boundary data through lifts and measure the resulting residual in an L2L^2 product space.

Let Ω⊂Rd\Omega\subset\mathbb{R}^d, d∈{2,3}d \in\{2,3\}, be a bounded Lipschitz domain. We partition its boundary into a Dirichlet part ΓD\Gamma_D of positive measure and a Neumann part ΓN=∂Ω∖ΓD\Gamma_N = \partial\Omega\setminus\Gamma_D. We use (⋅,⋅)(\cdot,\cdot) and ∥⋅∥\|\cdot\| for the L2(Ω)L^2(\Omega) inner product and norm, including their vector- and matrix-valued counterparts. All other norms carry a subscript. For scalar diffusion problems, we define

U=H0,ΓD1(Ω),Σ=H0,ΓN(div⁡;Ω):={τ∈H(div⁡;Ω):τ⋅n=0 on ΓN},U = H^1_{0,\Gamma_D}(\Omega), \qquad\Sigma= H_{0,\Gamma_N}(\operatorname{div};\Omega) := \{\tau\in H(\operatorname{div};\Omega) : \tau\cdot n = 0 \text{ on } \Gamma_N\},

where the boundary conditions are understood in the trace sense. We equip the state space H=Σ×U\mathbb{H} = \Sigma\times U with the PDE-compliant norm

∥[τ,v]∥H2=∥τ∥H(div⁡)2+∥v∥H12=∥τ∥2+∥div⁡τ∥2+∥v∥2+∥∇v∥2.(2)\|[\tau,v]\|_{\mathbb{H}}^2 = \|\tau\|_{H(\operatorname{div})}^2 + \|v\|_{H^1}^2 = \|\tau\|^2 + \|\operatorname{div}\tau\|^2 + \|v\|^2 + \|\nabla v\|^2. \tag*{(2)}

For the elasticity problem, we take U=H0,ΓD1(Ω;Rd)U = H^1_{0,\Gamma_D}(\Omega;\mathbb{R}^d) and let Σ\Sigma consist of matrix fields with rows in H(div⁡;Ω)H(\operatorname{div};\Omega) and vanishing traction on ΓN\Gamma_N. The norm is defined as in (2), with divergence taken row-wise. We impose no symmetry on the stress, since the first Piola–Kirchhoff stress is generally nonsymmetric.

We measure residuals in the self-dual product space

L2={L2(Ω)d×L2(Ω),for scalar diffusion,L2(Ω)d×d×L2(Ω)d,for elasticity.\mathbb{L}^2 = \begin{cases} L^2(\Omega)^d \times L^2(\Omega), & \text{for scalar diffusion},\\ L^2(\Omega)^{d\times d} \times L^2(\Omega)^d, & \text{for elasticity}. \end{cases}

Throughout, CPC_P denotes the Poincaré constant satisfying ∥v∥≤CP∥∇v∥\|v\| \le C_P \|\nabla v\| for v∈Uv \in U. We summarize the notation in Table 1.

References

  1. [1]L. Lu, P. Jin, G. Pang, Z. Zhang, G. E. Karniadakis, Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators, Nature Machine Intelligence 3 (3) (2021) 218–229.
  2. [2]Z. Li, N. Kovachki, K. Azizzadenesheli, B. Liu, K. Bhattacharya, A. Stuart, A. Anandkumar, Fourier neural operator for parametric partial differential equations, in: International Conference on Learning Representations, 2021.
  3. [3]D. Z. Huang, A. M. Stuart, Cost-accuracy trade-offs: Neural operator vs classical numerical solver, arXiv:2609.24021 (2026). doi:10.48550/arXiv.2609.24021.DOI
  4. [4]M. Raissi, P. Perdikaris, G. 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 (2019) 686–707.DOI
  5. [5]E. Kharazmi, Z. Zhang, G. E. Karniadakis, hp-VPINNs: Variational physics-informed neural networks with domain decomposition, Computer Methods in Applied Mechanics and Engineering 374 (2021) 113547.DOI
  6. [6]S. Wang, H. Wang, P. Perdikaris, Learning the solution operator of parametric partial differential equations with physics-informed DeepONets, Science Advances 7 (40) (2021) eabi8605. doi:10.1126/sciadv.abi8605.DOI
  7. [7]Z. Li, H. Zheng, N. Kovachki, D. Jin, H. Chen, B. Liu, K. Azizzadenesheli, A. Anandkumar, Physics-informed neural operator for learning partial differential equations, ACM/IMS Journal of Data Science 1 (3) (2024) 1–27. doi:10.1145/3648506.DOI
  8. [8]Z. Cai, R. Lazarov, T. A. Manteuffel, S. F. McCormick, First-order system least squares for second-order partial differential equations: Part I, SIAM Journal on Numerical Analysis 31 (6) (1994) 1785–1799. doi:10.1137/0731091.DOI
  9. [9]Z. Cai, T. A. Manteuffel, S. F. McCormick, First-order system least squares for second-order partial differential equations: Part II, SIAM Journal on Numerical Analysis 34 (2) (1997) 425–454. doi:10.1137/S0036142994266066.DOI
  10. [10]Z. Cai, G. Starke, First-order system least squares for the stress-displacement formulation: Linear elasticity, SIAM Journal on Numerical Analysis 41 (2) (2003) 715–730. doi:10.1137/S003614290139600X.DOI
  11. [11]P. B. Bochev, M. D. Gunzburger, Least-Squares Finite Element Methods, Vol. 166 of Applied Mathematical Sciences, Springer, New York, 2009. doi:10.1007/b13382.DOI
  12. [12]M. Bachmayr, W. Dahmen, M. Oster, Variationally correct neural residual regression for parametric PDEs: On the viability of controlled accuracy, IMA Journal of Numerical Analysis (2025). doi:10.1093/imanum/draf073.DOI
  13. [13]P. Cortés Castillo, W. Dahmen, J. Gopalakrishnan, DPG loss functions for learning parameter-to-solution maps by neural networks, arXiv preprint arXiv:2506.18773 (2025).DOI
  14. [14]Y. Qiu, W. Dahmen, P. Chen, Variationally correct operator learning: Reduced basis neural operator with a posteriori error estimation, Computer Methods in Applied Mechanics and Engineering 461 (2026) 119228.
  15. [15]K. Bhattacharya, B. Hosseini, N. B. Kovachki, A. M. Stuart, Model reduction and neural networks for parametric PDEs, The SMAI Journal of Computational Mathematics 7 (2021) 121–157.arxiv.org/abs/2005.03180
  16. [16]N. Dal Santo, S. Deparis, L. Pegolotti, Data driven approximation of parametrized PDEs by reduced basis and neural networks, Journal of Computational Physics 416 (2020) 109550. doi:10.1016/j.jcp.2020.109550.DOI
  17. [17]S. Fresca, L. Dedè, A. Manzoni, A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs, Journal of Scientific Computing 87 (2) (2021) 61. doi:10.1007/s10915-021-01462-7.DOI
  18. [18]F. Brezzi, J. Rappaz, P.-A. Raviart, Finite dimensional approximation of nonlinear problems. Part I: Branches of nonsingular solutions, Numerische Mathematik 36 (1) (1980) 1–25. doi:10.1007/BF01395985.DOI
  19. [19]K. Veroy, C. Prud’homme, A. T. Patera, Reduced-basis approximation of the viscous Burgers equation: rigorous a posteriori error bounds, Comptes Rendus Mathématique 337 (9) (2003) 619–624. doi:10.1016/j.crma.2003.09.023.DOI
  20. [20]A. Quarteroni, A. Manzoni, F. Negri, Reduced Basis Methods for Partial Differential Equations: An Introduction, Vol. 92 of Unitext, Springer, 2016.DOI
  21. [21]B. Haasdonk, Reduced basis methods for parametrized PDEs—a tutorial introduction for stationary and instationary problems, in: P. Benner, A. Cohen, M. Ohlberger, K. Willcox (Eds.), Model Reduction and Approximation: Theory and Algorithms, SIAM, Philadelphia, 2017, pp. 65–136. doi:10.1137/1.9781611974829.ch2.
  22. [22]F. M. Bersetche, J. P. Borthagaray, A deep first-order system least squares method for solving elliptic PDEs, Computers & Mathematics with Applications 129 (2023) 136–150.arxiv.org/abs/2204.07227
  23. [23]J. A. A. Opsochoor, P. C. Petersen, C. Schwab, First-order system least-squares neural networks, arXiv preprint arXiv:2409.20264 (2024). doi:10.48550/arXiv.2409.20264.
  24. [24]F. Bertrand, M. Brodbecк, T. Ricken, H. Schneider, Least-squares finite element methods for nonlinear problems: A unified framework, arXiv preprint arXiv:2503.18739 (2025).arxiv.org/abs/2503.18739
  25. [25]L. Ernst, N. Rekatsinas, K. Urban, A posteriori certification for physics-informed neural networks, arXiv preprint arXiv:2502.20336 (2025).
  26. [26]J. D. Toscano, D. T. Chen, V. Oommen, J. Darbon, G. E. Karniadakis, A variational framework for residual-based adaptivity in neural PDE solvers and operator learning, arXiv:2509.14198 (2025). doi:10.48550/arXiv.2509.14198.arxiv.org/abs/2509.14198
  27. [27]H. Zheng, Y. Chen, J. Han, Y. Yu, ReBaNO: Reduced basis neural operator mitigating generalization gaps and achieving discretization invariance, arXiv:2509.09611 (2025). doi:10.48550/arXiv.2509.09611.
  28. [28]N. Bouziani, D. A. Ham, Physics-driven machine learning models coupling PyTorch and Firedrake, in: ICLR 2023 Workshop on Physics for Machine Learning, 2023, arXiv:2303.06871.arxiv.org/abs/2303.06871
  29. [29]S. K. Mitusch, S. W. Funke, J. S. Dokken, dolphin-adjoint 2018.1: automated adjoints for FEniCS and Firedrake, Journal of Open Source Software 4 (38) (2019) 1292. doi:10.21105/joss.01292.DOI
  30. [30]J. Y. Lee, S. Ko, Y. Hong, Finite element operator network for solving elliptic-type parametric PDEs, SIAM Journal on Scientific ComputingArXiv:2308.04690 (2025).arxiv.org/abs/2308.04690
  31. [31]Y. Hong, S. Ko, J. Lee, Error analysis for finite element operator learning methods for solving parametric second-order elliptic PDEs, arXiv:2404.17868 (2024).arxiv.org/abs/2404.17868
  32. [32]D. Gilbarg, N. S. Trudinger, Elliptic Partial Differential Equations of Second Order, Classics in Mathematics, Springer, Berlin, 2001, reprint of the 1998 edition.
  33. [33]P. Grisvard, Elliptic Problems in Nonsmooth Domains, Vol. 24 of Monographs and Studies in Mathematics, Pitman, Boston, 1985.
  34. [34]J. E. Marsden, T. J. R. Hughes, Mathematical Foundations of Elasticity, Prentice-Hall, Englewood Cliffs, NJ, 1983, reprinted by Dover, New York, 1994.
  35. [35]P. G. Ciarlet, Mathematical Elasticity. Vol. I: Three-Dimensional Elasticity, Vol. 20 of Studies in Mathematics and its Applications, North-Holland, Amsterdam, 1988.
  36. [36]A. Cohen, R. DeVore, Approximation of high-dimensional parametric PDEs, Acta Numerica 24 (2015) 1–159. doi:10.1017/S0962492915000033.DOI
  37. [37]M. S. Alnæs, Ulf: a finite element form language, in: Automated solution of differential equations by the finite element method: the FEniCS Book, Springer, 2012, pp. 303–338.DOI
  38. [38]A. Logg, K.-A. Mardal, G. Wells, Automated solution of differential equations by the finite element method: The FEniCS book, Vol. 84, Springer Science & Business Media, 2012.DOI
  39. [39]I. A. Baratta, J. P. Dean, J. S. Dokken, M. Habera, J. S. Hale, C. N. Richardson, M. E. Rognes, M. W. Scroggs, N. Sime, G. N. Wells, DOLFINx: the next generation FEniCS problem solving environment, Preprint, doi:10.5281/zenodo.10447666 (2023).
  40. [40]S. Chaturantabut, D. C. Sorensen, Nonlinear model reduction via discrete empirical interpolation, SIAM Journal on Scientific Computing 32 (5) (2010) 2737–2764. doi:10.1137/090766498.DOI
  41. [41]C. Farhat, P. Avery, T. Chapman, J. Cortial, Dimensional reduction of nonlinear finite element dynamic models with finite rotations and energy-based mesh sampling and weighting for computational efficiency, International Journal for Numerical Methods in Engineering 98 (9) (2014) 625–662. doi:10.1002/nme.4668.DOI
  42. [42]U. Villa, N. Petra, O. Ghattas, hIPPYlib: An extensible software framework for large-scale inverse problems governed by PDEs: Part I: Deterministic inversion and linearized Bayesian inference, ACM Transactions on Mathematical Software (TOMS) 47 (2) (2021) 1–34.
  43. [43]A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, et al., PyTorch: An imperative style, high-performance deep learning library, in: Advances in Neural Information Processing Systems, Vol. 32, 2019.arxiv.org/abs/1912.01703
  44. [44]J. Kossaifi, N. Kovachki, Z. Li, D. Pitt, M. Liu-Schiaffini, R. J. George, B. Bonev, K. Azizzadenesheli, J. Berner, V. Duruissseaux, A. Anandkumar, A library for learning neural operators, arXiv preprint arXiv:2412.10354 (2025).arxiv.org/abs/2412.10354
  45. [45]A. Ganesh­ram, H. Maust, V. Duruissseaux, Z. Li, Y. Wang, D. Leibovici, O. Bruno, T. Hou, A. Anandkumar, FC-PINO: High precision physics-informed neural operators via Fourier continuation, arXiv preprint arXiv:2211.15960 (2022).arxiv.org/abs/2211.15960
  46. [46]D. Amsallem, M. J. Zahr, C. Farhat, Nonlinear model order reduction based on local reduced-order bases, International Journal for Numerical Methods in Engineering 92 (10) (2012) 891–916. doi:10.1002/nme.4371.DOI
  47. [47]K. Lee, K. T. Carlberg, Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders, Journal of Computational Physics 404 (2020) 108973. doi:10.1016/j.jcp.2019.108973.
  48. [48]K. Jordan, Y. Jin, V. Boza, J. You, F. Cesista, L. Newhouse, J. Bernstein, Muon: An optimizer for hidden layers in neural networks (2024). URL https://kellerjordan.github.io/posts/muon/

Paper details

vlnc:W_88sj967c
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents