Pith. sign in

REVIEW 2 major objections 8 minor 37 references

Kernel-based Operator Learning: Error Analysis, Budget Allocation, and a Physics-Informed Extension

T0 review · 2 major / 8 minor · reviewed 2026-07-08 · glm-5.2

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

Pith's one-line read Budget allocation rule for kernel operator learning

desk verdict Solid budget allocation theory for kernel operator learning, but experiments violate the theory's own regularity assumptions read the letter →

arxiv 2607.06287 v1 pith:PXHSQZ6K submitted 2026-07-07 math.NA cs.LGcs.NA

Rüdiger Kempf This is my paper · ORCID
classification math.NAcs.LGcs.NA
keywords operatorreconstructionlearningerrorkernelkernel-basedonlinephysics-informed
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 proves that a two-stage kernel surrogate for operator learning converges to the true operator at an explicit rate, provided the number of training pairs N, the number of input observation points n, and the output resolution m are coupled by a precise scaling law. The surrogate works in two stages: an offline kernel regression learns a finite-dimensional map from discretized inputs to discretized outputs, and an online kernel reconstruction recovers the output function from those predicted values. The author's key insight is to treat the online stage as a reconstruction from perturbed data, which splits the total error into a reconstruction term (governed by output grid density) and a learning term (governed by training data density), coupled only through a single regularization parameter. Balancing that parameter yields the budget allocation condition: the total error converges to zero as m and N grow, with the oracle rate m^{-(σ-τ)/d}, whenever log N / log m ≥ np(2σ-d) / [d(2α-np)]. If N scales exactly at this threshold, both error terms decay at the same rate and the learning error does not pollute the reconstruction rate. The paper also introduces a physics-informed variant: at evaluation time, the online reconstruction is augmented with a soft penalty on PDE residuals at collocation points, implemented via a closed-form representer theorem. This requires no retraining for new inputs and no full PDE solve, and numerically reduces reconstruction error below the data-driven baseline.

What carries the argument

Two-stage kernel surrogate A = A_on ∘ A_off ∘ S_X; error decomposition into reconstruction error (Term I) and learning error (Term II) coupled via regularization parameter λ; sampling inequalities bounding Sobolev norms by fill-distance-dependent terms; physics-informed Tikhonov functional with PDE residual penalty at collocation points

What would settle it

If one constructs an operator G and model class M where the discretized operator g either does not exist as an element of the assumed Sobolev RKHS or has substantially lower smoothness than α, the budget allocation condition would predict convergence that does not occur, and the error bounds of Theorems 24 and 28 would not hold.

Watch

Extended reading notes

Core claim

The central object is the budget allocation condition (Equation 10): log N / log m ≥ np(2σ−d) / [d(2α−np)]. This inequality is derived from a decomposition of the total surrogate error into a reconstruction error term scaling as m^{-(σ−τ)/d} and a learning error term scaling as N^{-(2α−np)/(2np)} · m^{(2σ−d)/(2d)}, with the two coupled through the regularization parameter λ. When N grows at the threshold rate, both terms match and the surrogate achieves the same convergence rate as reconstruction from exact data. The decomposition holds for both interpolation-based and regularized least-squares offline learning, with identical asymptotic rates after optimal parameter tuning.

Load-bearing premise

The entire error analysis depends on the existence of a function g living in a specific Sobolev-type reproducing kernel Hilbert space that exactly maps discretized inputs to discretized outputs. If the true operator G does not induce such a well-defined, sufficiently smooth finite-dimensional map, the convergence rates degrade or fail entirely.

Editorial extensions

If this is right

  • The budget allocation rule gives practitioners a concrete formula for sizing training datasets relative to output resolution, rather than relying on heuristics or empirical scaling experiments.
  • The asymmetry of the framework (no reconstruction needed on the input side) opens a direct path to kernel-based inverse problem surrogates, where the roles of input and output spaces are swapped.
  • The physics-informed reconstruction can be layered onto any already-trained kernel operator surrogate at deployment time, improving output fidelity without retraining or invoking a PDE solver.
  • The identical asymptotic rates for interpolation and regularized least-squares suggest that the choice between the two offline methods should be driven by numerical conditioning and noise robustness rather than convergence rate.

Reading between the lines

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

  • The budget allocation condition implies a curse of dimensionality in the input observation dimension n: the required N grows as m raised to a power proportional to np, so increasing the input resolution n demands disproportionately more training data. This suggests that PCA compression of inputs (as used in the experiments) is not merely convenient but may be essential for tractability.
  • The physics-informed extension's convergence rate is left unanalyzed. If the PDE residual penalty effectively increases the smoothness of the reconstruction (as the numerical experiments suggest), one might expect an improved effective σ in the reconstruction error term, potentially relaxing the budget allocation condition.
  • The saturation of the PI surrogate at the learning error of A_off (observed in the Poisson experiments) is a direct manifestation of the budget allocation theory: once reconstruction is accurate enough, the fixed learning error dominates, and further increasing m or m_L yields no benefit without also increasing N.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 8 minor

Summary. The paper studies kernel-based operator learning in a two-stage sampling framework (offline kernel regression + online kernel reconstruction). The main theoretical contribution is an explicit budget allocation condition (Eq. 10) relating the number of training pairs N, the number of input observations n, and the output resolution m, derived from a coupled error decomposition (Eq. 7) using sampling inequalities. A physics-informed extension augmenting the online reconstruction with a soft PDE collocation penalty is also introduced, with a representer theorem (Theorem 36). Numerical experiments on Darcy flow and Poisson equations are presented.

Significance. The budget allocation rule (Corollary 26, Eq. 10) is a useful, explicit contribution to the approximation-theoretic foundation of kernel-based operator learning, extending the framework of Batlle et al. (2024) by providing convergence rates directly in terms of N, n, and m. The asymmetric formulation removing the need for an input-space reconstruction operator A_le (Observations 1-3) is a clean simplification. The physics-informed representer theorem (Theorem 36) and its block linear system (Eq. 15) are well-derived and practically relevant. The error decomposition (7) and the optimal lambda choice (8) are clearly presented.

major comments (2)
  1. §4.1, Assumption 23(3) and §6.1: The budget allocation condition (10) requires alpha > n_p/2 (Assumption 23(3)), which is the standard Sobolev embedding condition for H^alpha(B) -> C^0(B) when dim(B) = n_p. In the Darcy flow experiments (§6.1), the bottom kernel is Matérn-3/2 (alpha=2) and inputs are reduced via PCA to n_PCA = 20 or 28 components. Whether n_p = n_PCA or n_p = 4096 (spatial grid points), the condition alpha > n_p/2 is massively violated (2 > 10 or 2 > 2048). This renders the exponent in condition (10) negative, making the condition vacuous. The paper claims to 'validate the theoretical findings' (abstract) and that 'theoretical predictions are validated on the Darcy flow benchmark' (§7), but the experiments operate in a regime where the theory's assumptions do not hold. The paper should either (a) explicitly acknowledge that the experiments are in a regime not covered by
  2. §4.3, Assumption 23(4): The existence of g in H_{K_b} satisfying g(S_X(u)) = S_Y(G(u)) for all u in M is load-bearing for all error bounds (Theorems 24, 28) and the budget allocation condition (10). Section 4.3 discusses this only for the finite-dimensional case M = B_R[0] cap U_{n_0}, where S_X is injective on U_{n_0}. However, the paper does not verify that g (which maps R^{n_p} to R^{m*ell}) has the required Sobolev regularity alpha > n_p/2 in H^alpha(B)^{m*ell}. Injectivity of S_X on M ensures g is well-defined as a function, but says nothing about its smoothness as an element of the RKHS. The paper should state clearly what regularity of G and M are needed to guarantee g in H^alpha(B)^{m*ell}, or at minimum flag this as an open assumption with a discussion of when it can be expected to hold.
minor comments (8)
  1. §6.1, Fig. 3: The reference slope is labeled '-1' but the caption states the theoretical rate is m^{-sigma/d} = m^{-1}. It would help to state the oracle rate formula explicitly in the figure caption for readers who skip the text.
  2. §6.1: The regularization schedule used is lambda(m) = m^{-5/2}, but the theoretical optimum (Eq. 8) gives lambda* proportional to h_{Y,D}^{sigma - d/2} which, for quasi-uniform Y in d=2 with sigma=2, gives lambda* ~ m^{-(sigma-d/2)/d} = m^{-1/2}. The paper notes lambda(m) = m^{-5/2} is 'well within the flat plateau' (Fig. 5), but the discrepancy between the theoretical optimum and the practical schedule should be discussed more explicitly.
  3. §5.1, Remark 37: The smoothness requirement sigma > 2*nu + d/2 is explained, but the Poisson experiment (§6.2) uses Matérn-9/2 (sigma=5.5) with nu=2, d=2, requiring sigma > 5. The choice sigma=5.5 barely satisfies this. This should be noted as a near-marginal case.
  4. §2.4, Theorem 2: The condition (1) is a Lipschitz stability condition for the inverse sampling map. It would help to state this interpretation in the theorem statement itself, not only in the subsequent paragraph, to aid readability.
  5. §3.3, Proposition 20: The proof references 'Theorem 20' but should reference 'Proposition 20'. Similarly, 'Theorem 14' should be 'Proposition 14' and 'Theorem 17' should be 'Proposition 17'.
  6. §4.1, Corollary 26: The condition (10) requires 2*alpha - n_p > 0 for the exponent to be positive. This is equivalent to Assumption 23(3) but should be stated explicitly in the corollary for clarity.
  7. §6.2, Fig. 6: The PI surrogate saturates at approximately 0.023 from m=121 onward. The caption attributes this to 'irreducible learning error of A_off in n_PCA=28 input dimensions.' A brief quantitative connection to the budget allocation analysis would strengthen this claim.
  8. The abstract states 'Numerical experiments illustrate the theoretical findings.' Given the regime mismatch in §6.1, this phrasing should be softened or qualified.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for a careful and substantive reading of our manuscript. Both major comments identify genuine gaps between the theory and the experiments/assumptions that we will address in revision. We respond to each below.

read point-by-point responses

  1. Referee: Major Comment 1: Assumption 23(3) requires α > n_p/2, but in the Darcy experiments (Matérn-3/2, α=2, n_PCA=20 or 28), this condition is massively violated, making condition (10) vacuous. The paper claims to 'validate theoretical findings' but experiments are outside the theory's regime.

    Authors: The referee is correct that Assumption 23(3) (α > n_p/2) is violated in the Darcy flow experiments. With α = 2 (Matérn-3/2) and n_p = n_PCA = 20 or 28, the condition 2 > n_p/2 fails, and the exponent (2α - n_p) in condition (10) becomes negative, rendering the budget allocation condition vacuous in this regime. We acknowledge this without reservation. We will revise the manuscript in the following ways: (1) We will add an explicit remark in Section 6.1 stating that the experiments do not satisfy Assumption 23(3) and that condition (10) is not directly applicable. (2) We will soften the claims in the abstract and Section 7 from 'validate the theoretical findings' to language that accurately reflects what is and is not validated. Specifically, the oracle convergence experiment (Fig. 3) validates the reconstruction rate m^{-(σ-τ)/d} = m^{-1}, which depends only on the output kernel smoothness σ and the spatial dimension d, not on α or n_p. This part of the theory is genuinely tested. The budget allocation experiment (Fig. 4) validates the qualitative prediction that N must grow superlinearly with m to avoid stagnation, but the specific threshold exponent in (10) is not tested because the assumptions underlying it are not met. We will state this distinction clearly. (3) We will add a discussion of what kernel smoothness would be required to satisfy α > n_p/2 in the experimental setting (e.g., α > 10 for n_PCA = 20, corresponding to Matérn kernels of sufficiently high order) and note that this is a practical limitation of high-dimensional kernel methods, not specific to our framework. revision: yes

  2. Referee: Major Comment 2: The existence of g in H_{K_b} with the required Sobolev regularity α > n_p/2 is load-bearing but not verified. Section 4.3 only shows g is well-defined (injectivity of S_X on M) but does not address its smoothness as an element of the RKHS.

    Authors: The referee is correct. Section 4.3 establishes that g is well-defined as a function (via injectivity of S_X on the finite-dimensional model class M), but we do not verify that g belongs to H^α(B)^{m·ℓ} with the required smoothness α > n_p/2. This is a genuine gap: injectivity ensures existence of g as a mapping, but says nothing about its Sobolev regularity. We will address this in revision as follows: (1) We will add an explicit remark after Assumption 23(4) flagging that the Sobolev regularity of g is an open assumption, distinct from its well-definedness. (2) We will add a discussion of settings where the regularity of g can be expected to hold. In the finite-dimensional case M = B_R[0] ∩ U_{n_0} with U_{n_0} finite-dimensional, if G: U → V is sufficiently smooth (e.g., G ∈ C^k(U, V) for k large enough) and S_X, S_Y are point evaluation operators, then g = S_Y ∘ G ∘ S_X^{-1} inherits smoothness from G and S_X^{-1}. Since S_X^{-1}: S_X(U_{n_0}) → U_{n_0} is a linear map between finite-dimensional spaces, g is as smooth as G composed with finite-dimensional linear maps, which for analytic or C^∞ operators G (as in the Darcy setting with smooth coefficients) yields g ∈ C^∞ and hence g ∈ H^α for any α. However, the quantitative requirement α > n_p/2 becomes increasingly stringent as n_p grows, and we will note this tension explicitly. (3) We will also note that in the mismatch framework of Remark 25, one can work with g ∈ H^β(B)^{m·ℓ} for β < α, at the cost of modified convergence rates, which partially mitigates the issue but does not eliminate the need for β > n_p/2. revision: yes

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: budget allocation rule derived from independent error analysis, not from fitted data or self-citation

full rationale

The paper's central result — the budget allocation condition (10) — is derived from a chain of independent, standard approximation-theoretic results, not from circular or self-referential logic. The derivation proceeds as follows: (1) The error decomposition (7) splits the total error into a reconstruction term (Term I) and a learning term (Term II) via the linearity of A_on = Q_{λ,Y}, using the operator norm bound ||A_on|| ≤ λ^{-1/2} from Proposition 17. (2) Term I is bounded by applying Proposition 20 (sampling inequalities from Gia et al. 2025, an independent external source) to the regularized least-squares operator. (3) Term II is bounded by combining the same Proposition 17 with Proposition 20 applied to the interpolation operator I_U. (4) The optimal λ* in Eq. (8) is obtained by balancing the two fill-distance-dependent exponents in f_1 and f_2 — a standard optimization step, not a fit to data. (5) Corollary 26 then substitutes quasi-uniform fill distance relations h_{Y,D} ~ m^{-1/d} and h_{U,B} ~ N^{-1/(np)} (Remark 22) into the balanced bound to obtain condition (10). Every step is a first-principles derivation from externally established results (Wendland 2004, Gia et al. 2025, Narcowich et al. 2005, Arcangeli et al. 2012). The self-citation to Batlle et al. (2024) is for the two-stage framework itself, not for any load-bearing theorem or uniqueness claim; the paper explicitly extends and relaxes assumptions from that work (Observations 1-3, Remark 31). The numerical experiments in Section 6 validate the scaling law m^{-1} but do not fit its constants — the rate is predicted from σ=2, d=2 and confirmed empirically. The regularization schedule λ(m) = m^{-5/2} is chosen to lie in the flat plateau, not fitted to match the theory's λ* = m^{-1}. The minor score of 2 reflects that the framework builds on Batlle et al. (2024) for its setting, but the error analysis and budget allocation rule are independently derived and do not reduce to that citation.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities or particles. The free parameters (λ, µ) are derived optimally from the error analysis rather than fitted, though the multiplicative constants c_λ, c_µ remain free. The PI weight µ_PDE is set heuristically in experiments without theoretical derivation. The axioms are standard domain assumptions in kernel approximation theory, with the existence of g being the most structurally important.

free parameters (3)
  • λ (online regularization) = λ* = c_λ (h_Y,D^{σ-d/2})^2
    Derived optimally by balancing reconstruction and learning error terms, not fitted to data. Constant c_λ is free.
  • µ (offline RLS regularization) = µ* = c_µ (h_U,B^{α-1/(2np)})^2
    Derived by balancing fill-distance and regularization bias terms. Constant c_µ is free.
  • µ_PDE (PI collocation weight) = µ = λ(m) in experiments
    Set equal to λ(m) in experiments; no theoretical derivation for optimal µ_PDE is provided.
assumptions (5)
  • domain assumption Existence of discretized operator g ∈ H_Kb with g(S_X(u)) = S_Y(G(u)) for all u ∈ M
    Assumption 23(4). Load-bearing for all error bounds. Discussed in Section 4.3 for finite-dimensional M.
  • domain assumption Lipschitz stability of inverse sampling map (Eq. 1)
    Theorem 2. Required to link fill distance in R^np to fill distance in U. Verified for finite-dimensional subspaces in Remark 32.
  • domain assumption Uniform a priori estimate for L_u over M
    Assumption 33(2). Required for the PI extension to ensure boundedness of the differential operator evaluation functionals.
  • standard math Sampling inequalities (Theorem 18)
    Cited from Gia et al. (2025). Standard tool in kernel approximation theory.
  • domain assumption Quasi-uniformity of point sets Y and U
    Required in Corollaries 26 and 30 to convert fill distances to sample sizes via h ~ M^{-1/d}.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Kernel-based Operator Learning: Error Analysis, Budget Allocation, and a Physics-Informed Extension." pith.science (2026). https://pith.science/paper/PXHSQZ6K

@misc{pith2026260706287,
  author       = {Pith},
  title        = {Pith review of: Kernel-based Operator Learning: Error Analysis, Budget Allocation, and a Physics-Informed Extension},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PXHSQZ6K}},
  note         = {Machine review of arXiv:2607.06287}
}
abstract

We study kernel-based operator learning in a two-stage sampling framework, where an offline kernel regression operator learns a discretized representation of the target operator from input-output pairs and an online kernel reconstruction operator recovers the output function from predicted observations. Our main theoretical contribution is an explicit budget allocation condition relating the number $N$ of training pairs, the number $n$ of input observations, and the output resolution $m$. The condition is derived from a coupled error analysis that interprets the surrogate as a reconstruction from approximate data. This yields a decomposition of the total error into reconstruction and learning contributions that can be analyzed independently. As a consequence, we obtain quantitative scaling laws describing how $N$, $n$, and $m$ must be coupled to guarantee convergence and to balance offline learning and online reconstruction errors. The resulting estimates extend previous analyses of kernel-based operator learning. We further introduce a physics-informed extension that incorporates knowledge of the underlying PDE at evaluation time. Rather than encoding constraints directly into the kernel, we augment the online reconstruction step by penalizing PDE residuals at collocation points. The method requires no retraining for new inputs. Numerical experiments illustrate the theoretical findings and demonstrate the effectiveness of the proposed physics-informed reconstruction strategy.

Figures

Figures reproduced from arXiv: 2607.06287 by the authors.

Figure 1
Figure 1. The fully symmetric learning approach of Batlle et al. (2024): To learn the [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Asymmetric diagrams addressing Observation 1 and Observation 2: only one [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Convergence of the oracle and full surrogate error as a function of output resolu [PITH_FULL_IMAGE:figures/full_fig_p024_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Budget allocation experiment with N = mκ and nPCA = 20. The curve with κ = 0.5 stagnates, while all curves with κ ≥ 1.0 converge at the oracle rate and are essentially indistinguishable from each other. Regularization sensitivity [PITH_FULL_IMAGE:figures/full_fig_p025…
Figure 5
Figure 5. Figure 5: Oracle error as a function of λ for fixed m = 225. The error is flat over more than ten orders of magnitude to the left of the theoretical optimum λ ∗ = 225−1 ≈ 4.44 × 10−3 , confirming robustness of the regularization schedule λ(m) = m−5/2 . used are again diagonal ke…
Figure 6
Figure 6. Figure 6: Oracle, PI oracle, and PI surrogate errors as a function of output resolution [PITH_FULL_IMAGE:figures/full_fig_p027_6.png]
Figure 7
Figure 7. Figure 7: Effect of the number of collocation points [PITH_FULL_IMAGE:figures/full_fig_p028_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    Arcangeli, M

    R. Arcangeli, M. Cruz Lopez de Silanes, and J. J. Torrens. Extension of sampling inequalities to S obolev semi-norms of fractional order and derivative data. Numer. Math., 121 0 (3): 0 587--608, 2012

    work page 2012

  2. [2]

    Kernel methods are competitive for operator learning

    Pau Batlle, Matthieu Darcy, Bamdad Hosseini, and Houman Owhadi. Kernel methods are competitive for operator learning. Journal of Computational Physics, 496: 0 112549, 2024. ISSN 0021-9991. doi:https://doi.org/10.1016/j.jcp.2023.112549. URL https://www.sciencedirect.com/science/article/pii/S0021999123006447

    work page doi:10.1016/j.jcp.2023.112549 2024

  3. [3]

    and Scott, L

    Susanne C. Brenner and L. Ridgway Scott. The Mathematical Theory of Finite Element Methods. Springer New York, 2008. ISBN 9780387759340. doi:10.1007/978-0-387-75934-0. URL http://dx.doi.org/10.1007/978-0-387-75934-0

    work page doi:10.1007/978-0-387-75934-0 2008

  4. [4]

    Yifan Chen, Bamdad Hosseini, Houman Owhadi, and Andrew M. Stuart. Solving and learning nonlinear pdes with gaussian processes. Journal of Computational Physics, 447: 0 110668, 2021. ISSN 0021-9991. doi:https://doi.org/10.1016/j.jcp.2021.110668. URL https://www.sciencedirect.com/science/article/pii/S0021999121005635

    work page doi:10.1016/j.jcp.2021.110668 2021

  5. [5]

    Sur l erreur d interpolation des fonctions de plusieurs variables par les d^m -splines

    Jean Duchon. Sur l erreur d interpolation des fonctions de plusieurs variables par les d^m -splines. RAIRO. Analyse num\'erique, 12 0 (4): 0 325--334, 1978. URL https://www.numdam.org/item/M2AN_1978__12_4_325_0/

    work page 1978

  6. [6]

    Partial Differential Equations

    Lawrence C Evans. Partial Differential Equations. Graduate Studies in Mathematics. American Mathematical Society, Providence, RI, 2 edition, March 2010

    work page 2010

  7. [7]

    Kernel-based Approximation Methods using MATLAB

    Gregory Fasshauer and Michael McCourt. Kernel-based Approximation Methods using MATLAB. WORLD SCIENTIFIC, June 2014. ISBN 9789814630146. doi:10.1142/9335. URL http://dx.doi.org/10.1142/9335

    work page doi:10.1142/9335 2014

  8. [8]

    Fuselier

    Edward J. Fuselier. Sobolev-type approximation rates for divergence-free and curl-free rbf interpolants. Mathematics of Computation, 77 0 (263): 0 1407--1423, 2008. ISSN 00255718, 10886842. URL http://www.jstor.org/stable/40234564

    work page arXiv 2008

Show all 37 references
  1. [9]

    Vector-valued gaussian processes for approximating divergence- or rotation-free vector fields, 2025

    Quoc Thong Le Gia, Ian Hugh Sloan, and Holger Wendland. Vector-valued gaussian processes for approximating divergence- or rotation-free vector fields, 2025. URL https://arxiv.org/abs/2511.12535

    2025

  2. [10]

    Operator-valued kernels for learning from functional response data

    Hachem Kadri, Emmanuel Duflos, Philippe Preux, St \'e phane Canu, Alain Rakotomamonjy, and Julien Audiffren. Operator-valued kernels for learning from functional response data. Journal of Machine Learning Research, 17 0 (20): 0 1--54, 2016. URL http://jmlr.org/papers/v17/11-315.html

    2016

  3. [11]

    E.J. Kansa. Multiquadrics—a scattered data approximation scheme with applications to computational fluid-dynamics—i surface approximations and partial derivative estimates. Computers and Mathematics with Applications, 19 0 (8-9): 0 127–145, 1990. ISSN 0898-1221. doi:10.1016/08...

    1990 doi

  4. [12]

    Stuart, and Anima Anandkumar

    Zongyi Li, Nikola Borislavov Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, Andrew M. Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differential equations. In 9th International Conference on Learning Representations, ICLR 2...

    2021

  5. [13]

    A kernel framework for learning differential equations and their solution operators

    Da Long, Nicole Mrvaljević, Shandian Zhe, and Bamdad Hosseini. A kernel framework for learning differential equations and their solution operators. Physica D: Nonlinear Phenomena, 460: 0 134095, 2024. ISSN 0167-2789. doi:https://doi.org/10.1016/j.physd.2024.134095. URL https:/...

    2024 doi

  6. [14]

    Learning nonlinear operators via deeponet based on the universal approximation theorem of operators

    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 0 (3): 0 218–229, March 2021. ISSN 2522-5839. doi:10.1038/s42256-0...

    2021 doi

  7. [15]

    A comprehensive and fair comparison of two neural operators (with practical extensions) based on fair data

    Lu Lu, Xuhui Meng, Shengze Cai, Zhiping Mao, Somdatta Goswami, Zhongqiang Zhang, and George Em Karniadakis. A comprehensive and fair comparison of two neural operators (with practical extensions) based on fair data. Computer Methods in Applied Mechanics and Engineering, 393: 0...

    2022 doi

  8. [16]

    Mat \'e rn

    B. Mat \'e rn. Spatial variation, volume 36 of Lecture Notes in Statistics. Springer, Berlin, second edition, 1986. ISBN 3-540-96365-0. doi:10.1007/978-1-4615-7892-5

    1986 doi

  9. [17]

    Micchelli and Massimiliano Pontil

    Charles A. Micchelli and Massimiliano Pontil. On learning vector-valued functions. Neural Computation, 17 0 (1): 0 177–204, January 2005. ISSN 1530-888X. doi:10.1162/0899766052530802. URL http://dx.doi.org/10.1162/0899766052530802

    2005 doi

  10. [18]

    Operator learning with gaussian processes

    Carlos Mora, Amin Yousefpour, Shirin Hosseinmardi, Houman Owhadi, and Ramin Bostanabad. Operator learning with gaussian processes. Computer Methods in Applied Mechanics and Engineering, 434: 0 117581, 2025. ISSN 0045-7825. doi:https://doi.org/10.1016/j.cma.2024.117581. URL htt...

    2025 doi

  11. [19]

    Narcowich and Joseph D

    Francis J. Narcowich and Joseph D. Ward. Generalized hermite interpolation via matrix-valued conditionally positive definite functions. Mathematics of Computation, 63 0 (208): 0 661–661, 1994. ISSN 0025-5718. doi:10.1090/s0025-5718-1994-1254147-6. URL http://dx.doi.org/10.1090...

    1994 doi

  12. [20]

    Narcowich, Joseph D

    Francis J. Narcowich, Joseph D. Ward, and Holger Wendland. Sobolev bounds on functions with scattered zeros, with applications to radial basis function surface fitting. Mathematics of Computation, 74 0 (250): 0 743--763, 2005

    2005

  13. [21]

    Sobolev error estimates and a bernstein inequality for scattered data interpolation via radial basis functions

    Francis J Narcowich, Joseph D Ward, and Holger Wendland. Sobolev error estimates and a bernstein inequality for scattered data interpolation via radial basis functions. Constructive Approximation, 24 0 (2): 0 175--186, 9 2006

    2006

  14. [22]

    Narcowich, Joseph D

    Francis J. Narcowich, Joseph D. Ward, and Grady B. Wright. Divergence-free rbfs on surfaces. Journal of Fourier Analysis and Applications, 13 0 (6): 0 643–663, October 2007. ISSN 1531-5851. doi:10.1007/s00041-006-6903-2. URL http://dx.doi.org/10.1007/s00041-006-6903-2

    2007 doi

  15. [23]

    Nelsen and Andrew M

    Nicholas H. Nelsen and Andrew M. Stuart. Operator learning using random features: A tool for scientific computing. SIAM Review, 66 0 (3): 0 535--571, 2024. doi:10.1137/24M1648703. URL https://doi.org/10.1137/24M1648703

    2024 doi

  16. [24]

    Do ideas have shape? idea registration as the continuous limit of artificial neural networks

    Houman Owhadi. Do ideas have shape? idea registration as the continuous limit of artificial neural networks. Physica D: Nonlinear Phenomena, 444: 0 133592, 2023. ISSN 0167-2789. doi:https://doi.org/10.1016/j.physd.2022.133592. URL https://www.sciencedirect.com/science/article/...

    2023 doi

  17. [25]

    Operator-Adapted Wavelets, Fast Solvers, and Numerical Homogenization: From a Game Theoretic Approach to Numerical Approximation and Algorithm Design

    Houman Owhadi and Clint Scovel. Operator-Adapted Wavelets, Fast Solvers, and Numerical Homogenization: From a Game Theoretic Approach to Numerical Approximation and Algorithm Design. Cambridge University Press, October 2019. ISBN 9781108484367. doi:10.1017/9781108594967. URL h...

    2019 doi

  18. [26]

    U ber die Zerlegung einer empirischen Funktion in Sinuswellen . Schl \

    C. Runge. \"U ber die Zerlegung einer empirischen Funktion in Sinuswellen . Schl \"o milch Z. , 52: 0 117--123, 1905

    1905

  19. [27]

    Unsymmetric meshless methods for operator equations

    Robert Schaback. Unsymmetric meshless methods for operator equations. Numerische Mathematik, 114 0 (4): 0 629–651, October 2009. ISSN 0945-3245. doi:10.1007/s00211-009-0265-z. URL http://dx.doi.org/10.1007/s00211-009-0265-z

    2009 doi

  20. [28]

    Kernel techniques: From machine learning to meshless methods

    Robert Schaback and Holger Wendland. Kernel techniques: From machine learning to meshless methods. Acta Numerica, 15: 0 543–639, 2006. doi:10.1017/S0962492906270016

    2006 doi

  21. [29]

    Bernhard Sch\" o lkopf and Alexander J. Smola. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. The MIT Press, December 2001. ISBN 9780262256933. doi:10.7551/mitpress/4175.001.0001. URL http://dx.doi.org/10.7551/mitpress/4175.001.0001

    2001 doi

  22. [30]

    Fluids you can trust: Property-preserving operator learning for incompressible flows, 2026

    Ramansh Sharma, Matthew Lowery, Houman Owhadi, and Varun Shankar. Fluids you can trust: Property-preserving operator learning for incompressible flows, 2026. URL https://arxiv.org/abs/2602.15472

    2026 arXiv

  23. [31]

    Support Vector Machines

    Ingo Steinwart and Andreas Christmann. Support Vector Machines. Springer New York, 2008. ISBN 9780387772424. doi:10.1007/978-0-387-77242-4. URL http://dx.doi.org/10.1007/978-0-387-77242-4

    2008 doi

  24. [32]

    Optimizing shape parameters in rbf methods: A systematic review of techniques, applications, and computational challenges

    Jian Sun and Wenshuai Wang. Optimizing shape parameters in rbf methods: A systematic review of techniques, applications, and computational challenges. Computer Science Review, 59: 0 100842, 2026. ISSN 1574-0137. doi:https://doi.org/10.1016/j.cosrev.2025.100842. URL https://www...

    2026 doi

  25. [33]

    Learning the solution operator of parametric partial differential equations with physics-informed DeepONets

    Sifan Wang, Hanwen Wang, and Paris Perdikaris. Learning the solution operator of parametric partial differential equations with physics-informed DeepONets . Science Advances, 7 0 (40): 0 eabi8605, 2021. doi:10.1126/sciadv.abi8605

    2021 doi

  26. [34]

    Improved architectures and training algorithms for deep operator networks

    Sifan Wang, Hanwen Wang, and Paris Perdikaris. Improved architectures and training algorithms for deep operator networks. Journal of Scientific Computing, 92 0 (2), 2022. ISSN 1573-7691. doi:10.1007/s10915-022-01881-0. URL http://dx.doi.org/10.1007/s10915-022-01881-0

    2022 doi

  27. [35]

    Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree

    Holger Wendland. Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree. Advances in Computational Mathematics, 4 0 (1): 0 389–396, December 1995. ISSN 1572-9044. doi:10.1007/bf02123482. URL http://dx.doi.org/10.1007/BF02123482

    1995 doi

  28. [36]

    Scattered Data Approximation

    Holger Wendland. Scattered Data Approximation. Cambridge Monographs on Applied and Computational Mathematics. Cambridge University Press, 2004. doi:10.1017/CBO9780511617539

    2004 doi

  29. [37]

    Approximate interpolation with applications to selecting smoothing parameters

    Holger Wendland and Christian Rieger. Approximate interpolation with applications to selecting smoothing parameters. Numerische Mathematik, 101 0 (4): 0 729–748, August 2005. ISSN 0945-3245. doi:10.1007/s00211-005-0637-y. URL http://dx.doi.org/10.1007/s00211-005-0637-y

    2005 doi

Pith tools

Reviewed July 8, 2026 · model on record in the stance chip above.