Pith. sign in

REVIEW 4 major objections 5 minor 47 references

Deep Semiparametric Partial Differential Equation Models

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper proves that a profiling M-estimator for semiparametric PDEs estimates unknown mechanisms at the optimal rate and physical parameters at parametric efficiency.

desk verdict A serious semiparametric theory paper whose efficiency claims rest on an unverified stability condition; worth a referee, not yet a cite. read the letter →

arxiv 2506.22779 v1 pith:LNOPPT44 submitted 2025-06-28 stat.ME

Ziyuan Chen , Shunxing Yan , Fang Yao This is my paper · ORCID
classification stat.ME MSC 62G0562G0862G2062F1268T07
keywords semiparametricPDEprofilingM-estimationneuralnetworksefficiencynonparametricconvergenceratediscoveryroot-nconsistency
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper introduces SemiPDE, a model class in which observed data follow $Y = u(X) + \varepsilon$ and $u$ solves a PDE made of two parts: a known physical law $L(u;\theta)$ with unknown parameters $\theta$, and a completely unknown mechanism $F(u)$. The paper proposes a penalized profiling M-estimator that replaces $u$ by the numerical solution $\tilde u(x;\theta,F)$, so statistical estimation is decoupled from PDE solving, and estimates $F$ in an overparametrized neural network space. The paper claims this is the first semiparametric inference theory for deep M-estimation in PDE models, with the unknown mechanism estimated at the optimal nonparametric rate $n^{-2s/(2(s+r)+d)}\log n$, the solution at a faster rate, and $\theta$ at root-$n$ asymptotic normality and parametric efficiency. The payoff is a way to identify interpretable physical parameters from noisy data while allowing unknown terms to be learned rather than assumed away. The cost is a load-bearing stability condition that the authors note remains open to verify for general SemiPDE models.

What carries the argument

The machinery is the penalized profiling M-estimator in equation (6), where $u(x)$ is replaced by the numerical solution $\tilde u(x;\theta,F)$ of the SemiPDE and the unknown operator $F$ is searched inside an overparametrized feedforward neural network space $S_n$ with penalty $\|\phi - \phi_0\|^2$. This decoupling keeps statistical error confined to the neural-network parameters and lets the PDE solver contribute high accuracy without adding noise. The theoretical load is carried by Assumption 3 (Lipschitz, stability with a positive identifiability margin, and sensitivity conditions on the solution map $(\theta,F)\mapsto u$), by Assumptions 5 and 6 (existence of a sufficiently smooth orthogonal direction $h_0$ making the score $\psi = \partial_\theta l - \partial_F l[h_0]$ orthogonal to the nuisance tangent), and by the gradient-flow analysis of Algorithm 1 that yields the nonparametric rates.

What would settle it

Simulate from a SemiPDE whose stability margin shrinks with $n$, for example by placing the unknown mechanism under an integral operator that nearly annihilates its effect on the solution, so that the constant $C_3$ in Assumption 3(b) tends to zero; under the paper's theory the $L^2$ error of $\hat f$ should still track $n^{-2s/(2(s+r)+d)}\log n$ and the Theorem 4 intervals should hold coverage, so a visible departure from either would falsify the central claim.

Watch

Extended reading notes

Core claim

The central claim is that profiling a numerical PDE solver inside M-estimation turns the SemiPDE model into a tractable semiparametric problem: Theorem 1 gives $\|\hat f - f_0\|_{L^2}^2 = O_p(n^{-2s/(2(s+r)+d)}\log n)$ for the unknown mechanism, matching the minimax rate for estimating an $r$-th derivative up to a logarithm; Theorem 2 gives the solution rate $n^{-2(s+r)/(2(s+r)+d)}\log n$; and Theorem 3 shows $n^{1/2}(\hat\theta - \theta_0) \to N(0,\Sigma)$, with $\Sigma = \Sigma_{\mathrm{eff}}$ when Assumption 6 holds, so $\hat\theta$ is semiparametrically efficient. This is achieved without constructing an efficient score function: only the existence of a smooth direction $h_0$ annihilating the nuisance gradient is assumed, and the efficient covariance is reformulated as the minimizer of a convex optimization problem and estimated by finite differences. The paper therefore claims the first systematic analysis of neural-network estimators for PDE models that covers both the training dynamics, analyzed as gradient flow, and the differential structure of the model.

Load-bearing premise

The argument stands on Assumption 3: the mapping from the physical parameters and the unknown mechanism to the PDE solution must be Lipschitz, stable with a positive identifiability margin, and sensitive to $\theta$, uniformly over the parameter set; the paper itself flags that verifying these conditions for specific SemiPDE models remains open.

Editorial extensions

If this is right

  • If the theorems hold, practitioners can estimate the parametric part of a PDE at root-$n$ parametric efficiency without deriving an explicit efficient score function, which is unavailable for most PDE models.
  • The unknown mechanism $F$ is learned at the optimal nonparametric rate up to a log factor, so adding a data-driven term does not destroy the parametric convergence of $\theta$.
  • The profiling strategy avoids the slow convergence of joint PINN-style M-estimation, so PDE discovery can exploit accurate numerical solvers rather than treating the solution as a trainable function.
  • The finite-difference variance estimator in equations (13)-(14) turns efficient covariance estimation into a stable convex program, yielding valid simultaneous confidence intervals for linear functionals of $\theta$.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the decoupling idea is not specific to PDEs; the same profiling with a numerical forward map could extend semiparametric efficiency to other implicit models such as ODE systems and integral equations, provided stability and margin conditions analogous to Assumption 3 hold.
  • Editorial inference: the paper's rate for $\hat f$ degrades as $r$ grows, suggesting that high-order PDEs are inherently harder for mechanism recovery; a testable extension is to check whether a spectral or multi-scale parameterization of $F$ recovers faster rates than the neural-network space.
  • Editorial inference: the reliance on Gaussian noise in Assumption 6 means the parametric efficiency claim may not extend to heavy-tailed errors; replacing Gaussianity with sub-exponential tails and a robustified loss would be a natural stress test.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper introduces a semiparametric PDE (SemiPDE) model that combines a parametric physical operator L(u;θ) with a nonparametric unknown mechanism F(u), and proposes a profiling penalized M-estimation procedure in which the solution u is replaced by a numerical PDE solve for each (θ,F), with F represented by an overparameterized feedforward neural network. The main theoretical claims are (i) Theorem 1: an L2 convergence rate for the estimated mechanism f̂ of order n^{-2s/(2(s+r)+d)} log n under Assumptions 1-4; (ii) Theorem 2: a faster rate for the estimated solution û; (iii) Theorem 3: root-n consistency, asymptotic normality, and—under Assumption 6—semiparametric efficiency of the estimator θ̂ of the parametric part; and (iv) Theorem 4: an asymptotically valid confidence interval based on a numerically estimated efficient covariance. The paper also presents simulation comparisons against parametric, nonparametric, and PINN baselines, and two real-data applications (cell scratch assays and wave propagation through vegetation). The central mechanism is a set of Lipschitz, stability, and sensitivity conditions on the PDE solution map (Assumption 3), which the paper explicitly states remains open to verify for general SemiPDE models.

Significance. If the theoretical results are correct and the key assumptions are verifiable for practically relevant PDEs, this would be a substantial contribution: it would be the first semiparametric efficiency result for deep M-estimation in PDE settings, and it offers a concrete algorithmic template (profiling with numerical solvers) that avoids explicit score construction. The paper is also honest about the difficulty of its core assumptions, and it provides a useful taxonomy of how the PDE smoothing order r enters the nonparametric rate. However, the significance is currently conditional: the headline theorem (Theorem 3) rests on Assumption 3, which is not verified for any of the PDEs used in the simulations or real-data analyses, and the proofs are relegated to a supplementary file that is not part of the arXiv version under review. The paper's own Section 3.1 acknowledges that 'verifying these conditions for specific SemiPDE models remains an open and challenging problem,' and the only worked example (Example 3, Eq. (10)) is a first-order linear ODE, not a PDE.

major comments (4)
  1. [§3.1, Assumption 3 and Example 3 (Eq. (10))] Assumption 3 is the premise for all four theorems, but it is not verified for any PDE example in the paper. The only concrete verification, Example 3, is a first-order linear ODE (B_x u = θu + F(u) on [0,1]), not a partial differential equation. The reaction-diffusion, Nernst-Planck, and Navier-Stokes cases used in Section 5 are not analyzed with respect to the Lipschitz, stability, and sensitivity conditions. Since parabolic/elliptic solution maps can be smoothing in ways that may violate the lower bound in Eq. (9), the claimed rates in Theorems 1-3 would break down if the margin condition fails. The paper's own statement in Section 3.1 that verifying these conditions remains open is therefore a critical gap: the central claims are conditional on an assumption that is not demonstrated for any example in the manuscript.
  2. [§3.1, paragraph after Eq. (9), r = r1 restriction] The paper restricts the main theory to the case r = r1, asserting that this is 'commonly satisfied,' but no justification is provided for the reaction-diffusion, Nernst-Planck, or Navier-Stokes models. The smoothing order of the PDE solution map (which determines the appropriate r in the Lipschitz condition (8)) and the integrability order appearing in the stability condition (9) may differ; for parabolic equations the forward map is typically smoothing, while the inverse map may have a different order. Without an argument that r = r1 holds for at least one of the complex models in the headline claim, the rates in Theorems 1-2 and the efficiency result in Theorem 3 are not connected to the simulations.
  3. [Theorems 1-4 and Section 7 (Discussion)] The proofs of all theorems are in a supplementary file that is not included in the arXiv version under review, so the derivation cannot be independently checked. Additionally, the paper's final discussion (Section 7) lists 'verifying the critical though challenging assumptions, such as Lipschitz and stability conditions' as future work, which is a direct admission that the assumptions are not currently established for the models that the paper claims to cover. For a methodological paper whose main contribution is theoretical, the assumptions should be verified at least for a nontrivial PDE example—for instance, a reaction-diffusion equation with a known reaction function—or the claims should be restated as conditional on Assumption 3 without applying them to the simulated cases.
  4. [§4, Assumption 6 and Theorem 3] The efficiency claim in Theorem 3 requires Assumption 6, which imposes Gaussian noise, a specific form h0 = -B_θ f(V; θ0), and the identification of f(V;θ) as the population projection. This is a strong assumption that is not verified for the SemiPDE models in the simulations. While the paper notes that Theorem 3 remains value as a root-n consistency result without Assumption 6, the abstract's statement that 'our estimator for θ attains parametric efficiency for complex PDE models' is not supported unless Assumption 6 is shown to hold for such models. Please provide at least a characterization of which PDE families satisfy Assumption 6.
minor comments (5)
  1. [Throughout] Equation (5) uses 'Fpuq' as a differential operator but later the argument is 'Fpu(x))'; the notation should be made uniform, and the distinction between the operator F and its function representation f(V) should be clarified at first use.
  2. [§2.3, Eq. (7)] The definition of the FNN space uses L layers but the formula writes L_0 through L_{L+1}; the indexing is confusing and should be reconciled with the width vector dimensions m_0,...,m_{L+1}.
  3. [§5, Table 1] Table 1 would benefit from a column header indicating the quantity reported (e.g., L2 error versus parameter error) and from a statement of the Monte Carlo standard error, since the comparison between SemiPDE and Baseline 3 is sometimes close (e.g., Case 1 at n=800).
  4. [Algorithm 1 and §3.2] The stopping time T in Section 3.2 is defined via the gradient flow, but Algorithm 1 uses a validation-loss-based early stopping; the connection between the two stopping rules should be explained, otherwise it is unclear that the algorithmic estimator analyzed in the theory is the one implemented.
  5. [§6.1, Table 4] The estimates of θ for SemiPDE 1 (1232) and Benchmark 1 (853) are listed without standard errors or confidence intervals; given the paper's emphasis on inference, reporting the proposed confidence intervals for θ in the real-data example would strengthen the practical message.

Circularity Check

2 steps flagged · score 4.0 of 10

Efficiency proof leans on the authors' own prior tangent-space lemma, and Algorithm 1 is written to initialize at the true parameter; the core nonparametric rates are otherwise derived, not fitted.

  1. self citation load bearing [Section 4.1, paragraph immediately before Theorem 3]
    "In Yan et al. (2025), this issue is also identified as a general problem encountered in semiparametric M-estimation using neural networks. Adopting similar techniques in Yan et al. (2025) to show that h0 can be sufficiently approximated by functions in T_{\hat\phi}S_n, we need to address additional challenges caused by lack of a closed-form solution and the complex differential structures of the SemiPDE model."

    Theorem 3's root-n normality and efficiency conclusion requires that the empirical average of the orthogonal score psi be negligible, which in turn requires h0 to be approximated well by the neural-network tangent space T_{\hat\phi}S_n. The manuscript does not prove this approximation here; it says it adopts similar techniques from Yan et al. (2025), a preprint whose authors (S. Yan, Z. Chen, F. Yao) overlap with the present paper. Thus a load-bearing step in the central efficiency claim is justified by the authors' own prior work rather than by an independent theorem or by a proof contained in this paper.

  2. other [Algorithm 1, first line (Section 2.3)]
    "For k=0, choose initial parameters θ^{(0)} = θ0, weights ϕ^{(0)} = ϕ0 and Loss min = Inf;"

    In Section 3.1, θ0 is defined as the true parameter of the governing model. As written, the estimation algorithm therefore initializes at the true value θ0, so the simulation results in Section 5 and the claimed empirical support for the efficiency theory are obtained with the target parameter supplied as an algorithmic input. A user cannot supply θ0 before estimating it. If the notation is intended to mean a generic initial value rather than the true value, the manuscript never defines a distinct symbol such as θ_init, and the numerical experiments are not presented as a clean, initialization-free evaluation.

full rationale

The core derivation in Theorems 1 and 2 is a standard semiparametric M-estimation and gradient-flow argument: the rates n^{-2s/(2(s+r)+d)} log n and n^{-2(s+r)/(2(s+r)+d)} log n follow from Assumptions 1-4, the chosen penalty λ_n, and the Lipschitz/stability inequalities, which are substantive assumptions rather than fitted constants. The comparison with the known derivative-estimation minimax rate is an external consistency check, not a renaming. The main circularity concerns are the load-bearing reliance on Yan et al. (2025) for the tangent-space approximation needed in Theorem 3, and the algorithm's apparent initialization at the true θ0, which affects the reported numerical support. Assumption 3 being unverified for the complex PDE examples is a correctness and scope risk, but it is not itself a circularity because the theorems honestly state it as a premise. Overall, the central theoretical claims retain independent content, so a moderate score is appropriate.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the stability and identifiability conditions in Assumption 3 and the orthogonality/direction conditions in Assumptions 5-6. These are not independently verified for general PDEs and are essentially assumed to make the theory work. The free parameters are hyperparameters of the estimation algorithm whose theoretical values depend on unknown smoothness parameters.

free parameters (4)
  • penalty tuning parameter λ_n = chosen by validation in practice; theory sets λ_n = n^{-2(snn+r)/(2(s+r)+d)} log^{(snn+r)/(s+r)} n
    Controls the strength of the neural network penalty; its theoretical value depends on the unknown smoothness s and PDE order r.
  • neural network width m = chosen by hand, e.g., [16,64,64,16] in simulations; theory requires m ≥ (η/τ)^{6C}
    Width of the overparameterized FNN; must be large enough for the training dynamics theorem to hold.
  • learning rate η and stopping tolerance τ = η small, τ chosen so that τ/η ≤ n^{-2(s+r)/(2(s+r)+d)}
    The gradient flow approximation to Algorithm 1 requires a rate condition linking the optimization tolerance to the sample size and unknown smoothness.
  • numerical solver tolerance ε_u = ε_u ≤ (τ/η)^C
    Accuracy of the numerical PDE solution; must be smaller than the statistical tolerance.
assumptions (6)
  • domain assumption Assumption 1: θ0 belongs to a tight subset Θ of R^p and f0 belongs to Sobolev space W^{s,2} for some s>0.
    Standard regularity conditions for the parametric and nonparametric components.
  • domain assumption Assumption 2: The FNN space Sn has fixed depth, bounded width ratio, and a specific initialization scheme.
    Technical conditions on the neural network architecture required for the approximation and tangent space analysis.
  • ad hoc to paper Assumption 3: Lipschitz, stability (margin), and sensitivity conditions on the PDE solution map u(θ,F).
    Central identifiability and margin conditions; the paper admits verifying them for general SemiPDE models is open, and only gives examples for a linear equation and reaction-diffusion systems.
  • domain assumption Assumption 4: The numerical solver is accurate to within ε_u for all θ and F in the relevant sets.
    Assumes the forward PDE solver introduces negligible error relative to the statistical precision.
  • ad hoc to paper Assumption 5: There exists a smooth h0 in W^{s1,2} satisfying the orthogonality condition (11).
    Existence of an orthogonal direction for the nuisance parameter; essential for root-n normality and efficiency, but not verified for general PDE models.
  • ad hoc to paper Assumption 6: h0 = -Bθ f(V;θ0) and the noise is Gaussian.
    Strengthens Assumption 5 to guarantee efficiency; the Gaussian noise assumption is restrictive.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Deep Semiparametric Partial Differential Equation Models." pith.science (2026). https://pith.science/paper/LNOPPT44

@misc{pith2026250622779,
  author       = {Pith},
  title        = {Pith review of: Deep Semiparametric Partial Differential Equation Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LNOPPT44}},
  note         = {Machine review of arXiv:2506.22779}
}
read the original abstract

In many scientific fields, the generation and evolution of data are governed by partial differential equations (PDEs) which are typically informed by established physical laws at the macroscopic level to describe general and predictable dynamics. However, some complex influences may not be fully captured by these laws at the microscopic level due to limited scientific understanding. This work proposes a unified framework to model, estimate, and infer the mechanisms underlying data dynamics. We introduce a general semiparametric PDE (SemiPDE) model that combines interpretable mechanisms based on physical laws with flexible data-driven components to account for unknown effects. The physical mechanisms enhance the SemiPDE model's stability and interpretability, while the data-driven components improve adaptivity to complex real-world scenarios. A deep profiling M-estimation approach is proposed to decouple the solutions of PDEs in the estimation procedure, leveraging both the accuracy of numerical methods for solving PDEs and the expressive power of neural networks. For the first time, we establish a semiparametric inference method and theory for deep M-estimation, considering both training dynamics and complex PDE models. We analyze how the PDE structure affects the convergence rate of the nonparametric estimator, and consequently, the parametric efficiency and inference procedure enable the identification of interpretable mechanisms governing data dynamics. Simulated and real-world examples demonstrate the effectiveness of the proposed methodology and support the theoretical findings.

Figures

Figures reproduced from arXiv: 2506.22779 by the authors.

Figure 1
Figure 1. The estimation procedure of the SemiPDE model. [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. The distribution histogram of ˆθi ´ θ0. The red line is the density function of normal distribution with mean Bias and standard deviation Std, and the green dash line is Bias. Finally, we examine the identifiability and robustness of the proposed SemiPDE framework. In Cases 1 and 2, we assume known mechanism, i.e., Fpuq “ 0, meaning the true model is located within the parametric model family Lpu, θq “ 0. We compare… view at source ↗
Figure 3
Figure 3. The estimated reaction mechanism fpuq for each model. The estimates of θ in our two SemiPDE models are significantly larger than those from the benchmark models, yet they still fall within the range reported in Liang et al. (2007). Regarding the unknown mechanism, the reaction term fpuq estimated in our SemiPDE models initially increases as the density u grows from zero, but then decreases as u approaches to the max… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Boxplot for estimated CD in different environments. parametric PDE model, suffers significantly from model misspecification due to the complexity of real-world environments, leading to poor estimates of CD. In contrast, our model achieves higher accuracy, aided by mech…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 44 canonical work pages

  1. [1]

    Berg and K

    J. Berg and K. Nystr \"o m. Data-driven discovery of pdes in complex datasets. Journal of Computational Physics, 384: 0 239--252, 2019

    work page 2019

  2. [2]

    J. Blazek. Computational fluid dynamics: principles and applications. Butterworth-Heinemann, 2015

    work page 2015

  3. [3]

    J. Cao, J. Z. Huang, and H. Wu. Penalized nonlinear least squares estimation of time-varying parameters in ordinary differential equations. Journal of computational and graphical statistics, 21 0 (1): 0 42--56, 2012

    work page 2012

  4. [4]

    Chernozhukov, D

    V. Chernozhukov, D. Chetverikov, M. Demirer, E. Duflo, C. Hansen, W. Newey, and J. Robins. Double/debiased machine learning for treatment and structural parameters, 2018

    2018

  5. [5]

    W. Dai, T. Tong, and M. G. Genton. Optimal estimation of derivatives in nonparametric regression. Journal of Machine Learning Research, 17 0 (164): 0 1--25, 2016

    work page 2016

  6. [6]

    Dai and L

    X. Dai and L. Li. Kernel ordinary differential equations. Journal of the American Statistical Association, 117 0 (540): 0 1711--1725, 2022

    work page 2022

  7. [7]

    De Brabanter, J

    K. De Brabanter, J. De Brabanter, B. De Moor, and I. Gijbels. Derivative estimation with local polynomial fitting. Journal of Machine Learning Research, 14 0 (1), 2013

    work page 2013

  8. [8]

    De Ryck and S

    T. De Ryck and S. Mishra. Numerical analysis of physics-informed neural networks and related models in physics-informed machine learning. Acta Numerica, 33: 0 633--713, 2024

    work page 2024

Show all 47 references
  1. [9]

    M. Du, Y. Chen, and D. Zhang. Discover: Deep identification of symbolically concise open-form partial differential equations via enhanced reinforcement learning. Physical Review Research, 6 0 (1): 0 013182, 2024

    2024

  2. [10]

    M. B. Giles and N. A. Pierce. An introduction to the adjoint approach to design. Flow, turbulence and combustion, 65: 0 393--415, 2000

    2000

  3. [11]

    D. D. Ho, A. U. Neumann, A. S. Perelson, W. Chen, J. M. Leonard, and M. Markowitz. Rapid turnover of plasma virions and cd4 lymphocytes in hiv-1 infection. Nature, 373 0 (6510): 0 123--126, 1995

    1995

  4. [12]

    Z. Hu, S. Lian, H. Wei, Y. Li, M. Stive, and T. Suzuki. Laboratory data on wave propagation through vegetation with following and opposing currents. Earth System Science Data, 13 0 (10): 0 4987--4999, 2021

    2021

  5. [13]

    V. Isakov. Inverse problems for partial differential equations, volume 127. Springer, 2006

    2006

  6. [14]

    Jeng, J.-H

    D.-S. Jeng, J.-H. Ye, J.-S. Zhang, and P.-F. Liu. An integrated model for the wave-induced seabed response around marine structures: Model verifications and applications. Coastal Engineering, 72: 0 1--19, 2013

    2013

  7. [15]

    J. Jia, Y. Wu, P. Li, and D. Meng. Variational inverting network for statistical inverse problems of partial differential equations. Journal of Machine Learning Research, 24 0 (201): 0 1--60, 2023

    2023

  8. [16]

    W. Jin, E. T. Shah, C. J. Penington, S. W. McCue, L. K. Chopin, and M. J. Simpson. Reproducibility of scratch assays is affected by the initial degree of confluence: experiments, modelling and model selection. Journal of theoretical biology, 390: 0 136--145, 2016

    2016

  9. [17]

    C. Johnson. Numerical solution of partial differential equations by the finite element method. Courier Corporation, 2009

    2009

  10. [18]

    Kaltenbacher and W

    B. Kaltenbacher and W. Rundell. The inverse problem of reconstructing reaction--diffusion systems. Inverse Problems, 36 0 (6): 0 065011, 2020

    2020

  11. [19]

    Liang, A

    C.-C. Liang, A. Y. Park, and J.-L. Guan. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nature protocols, 2 0 (2): 0 329--333, 2007

    2007

  12. [20]

    Liang and H

    H. Liang and H. Wu. Parameter estimation for differential equation models using a framework of measurement error in regression models. Journal of the American Statistical Association, 103 0 (484): 0 1570--1583, 2008

    2008

  13. [21]

    L. Lu, X. Meng, Z. Mao, and G. E. Karniadakis. Deepxde: A deep learning library for solving differential equations. SIAM review, 63 0 (1): 0 208--228, 2021

    2021

  14. [22]

    Marmanis, C

    H. Marmanis, C. Hamman, and R. Kirby. A one-dimensional model of the navier-stokes. Technical report, Citeseer, 2006

    2006

  15. [23]

    R. Nickl. On bayesian inference for some statistical inverse problems with partial differential equations. Bernoulli News, 24 0 (2): 0 5--9, 2017

    2017

  16. [24]

    R. Nickl. Bayesian non-linear statistical inverse problems. 2023

    2023

  17. [25]

    Pantano, M

    A. Pantano, M. C. Boyce, and D. M. Parks. Nonlinear structural mechanics based modeling of carbon nanotube deformation. Physical review letters, 91 0 (14): 0 145504, 2003

    2003

  18. [26]

    Podina, B

    L. Podina, B. Eastman, and M. Kohandel. Universal physics-informed neural networks: symbolic differential operator discovery with sparse data. In International Conference on Machine Learning, pages 27948--27956. PMLR, 2023

    2023

  19. [27]

    Rackauckas, Y

    C. Rackauckas, Y. Ma, J. Martensen, C. Warner, K. Zubov, R. Supekar, D. Skinner, A. Ramadhan, and A. Edelman. Universal differential equations for scientific machine learning. arXiv preprint arXiv:2001.04385, 2020

    2001 arXiv

  20. [28]

    Raissi, P

    M. Raissi, P. Perdikaris, and G. 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: 0 686--707, 2019

    2019

  21. [29]

    J. O. Ramsay, G. Hooker, D. Campbell, and J. Cao. Parameter estimation for differential equations: a generalized smoothing approach. Journal of the Royal Statistical Society Series B: Statistical Methodology, 69 0 (5): 0 741--796, 2007

    2007

  22. [30]

    B. P. Rasmussen. Dynamic modeling for vapor compression systems—part i: Literature review. HVAC&R Research, 18 0 (5): 0 934--955, 2012

    2012

  23. [31]

    Sakthivel, S

    K. Sakthivel, S. Gnanavel, N. B. Balan, and K. Balachandran. Inverse problem for the reaction diffusion system by optimization method. Applied mathematical modelling, 35 0 (1): 0 571--579, 2011

    2011

  24. [32]

    Schmidt-Hieber

    J. Schmidt-Hieber. Nonparametric regression using deep neural networks with relu activation function. The Annals of Statistics, 48 0 (4): 0 1875, 2020

    2020

  25. [33]

    Sgura, A

    I. Sgura, A. S. Lawless, and B. Bozzini. Parameter estimation for a morphochemical reaction-diffusion model of electrochemical pattern formation. Inverse Problems in Science and Engineering, 27 0 (5): 0 618--647, 2019

    2019

  26. [34]

    Shi and R

    J. Shi and R. Shivaji. Persistence in reaction diffusion models with weak allee effect. Journal of Mathematical Biology, 52 0 (6): 0 807--829, 2006

    2006

  27. [35]

    Singh and S

    S. Singh and S. Saha Ray. Analysis of stochastic fitzhugh--nagumo equation for wave propagation in a neuron arising in certain neurobiology models. International Journal of Biomathematics, 15 0 (05): 0 2250027, 2022

    2022

  28. [36]

    J. M. Stockie. The mathematics of atmospheric dispersion modeling. Siam Review, 53 0 (2): 0 349--372, 2011

    2011

  29. [37]

    C. J. Stone. Optimal rates of convergence for nonparametric estimators. The annals of Statistics, pages 1348--1360, 1980

    1980

  30. [38]

    J. Tan, G. Zhang, X. Wang, H. Huang, and F. Yao. Green’s matching: an efficient approach to parameter estimation in complex dynamic systems. Journal of the Royal Statistical Society Series B: Statistical Methodology, page qkae031, 2024

    2024

  31. [39]

    Tenachi, R

    W. Tenachi, R. Ibata, and F. I. Diakogiannis. Deep symbolic regression for physics guided by units constraints: toward the automated discovery of physical laws. The Astrophysical Journal, 959 0 (2): 0 99, 2023

    2023

  32. [40]

    Vadeboncoeur, \"O

    A. Vadeboncoeur, \"O . D. Akyildiz, I. Kazlauskaite, M. Girolami, and F. Cirak. Fully probabilistic deep models for forward and inverse problems in parametric pdes. Journal of Computational Physics, 491: 0 112369, 2023

    2023

  33. [41]

    L. Wang, J. Cao, J. O. Ramsay, D. Burger, C. Laporte, and J. K. Rockstroh. Estimating mixed-effects differential equation models. Statistics and Computing, 24: 0 111--121, 2014

    2014

  34. [42]

    Y. Wang, J. Shi, and J. Wang. Persistence and extinction of population in reaction--diffusion--advection model with strong allee effect growth. Journal of mathematical biology, 78: 0 2093--2140, 2019

    2019

  35. [43]

    Wu and A

    H. Wu and A. A. Ding. Population hiv-1 dynamics in vivo: applicable models and inferential tools for virological data from aids clinical trials. Biometrics, 55 0 (2): 0 410--418, 1999

    1999

  36. [44]

    X. Xun, J. Cao, B. Mallick, A. Maity, and R. J. Carroll. Parameter estimation of partial differential equation models. Journal of the American Statistical Association, 108 0 (503): 0 1009--1020, 2013

    2013

  37. [45]

    S. Yan, Z. Chen, and F. Yao. Semiparametric m-estimation with overparameterized neural networks. arXiv preprint arXiv:2504.19089, 2025

    2025 arXiv

  38. [46]

    S. Yang, S. W. Wong, and S. Kou. Inference of dynamic systems from noisy and sparse data via manifold-constrained gaussian processes. Proceedings of the National Academy of Sciences, 118 0 (15): 0 e2020397118, 2021

    2021

  39. [47]

    Y. Yin, V. Le Guen, J. Dona, I. Ayed, E. de B \'e zenac, N. Thome, and P. Gallinari. Augmenting physical models with deep networks for complex dynamics forecasting. In Ninth International Conference on Learning Representations ICLR 2021, 2021

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.