Pith. sign in

REVIEW 3 major objections 4 minor 140 references

Mass Lumping and Numerical Quadrature for Approximation of Fractional Elliptic Differential Equations Driven by Gaussian White Noise

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Mass lumping and quadrature can replace exact finite-element matrices in fractional SPDE covariance approximations without reducing the convergence rate.

desk verdict Proves mass lumping preserves covariance rates in the SPDE setting, but the surface claims hinge on an unproved theorem, so the Euclidean and graph results deserve a referee while the manifold coverage is conditional. read the letter →

arxiv 2608.08658 v1 pith:IDGCZ3LX submitted 2026-08-09 math.NA cs.NAmath.STstat.TH

Kelvin J. R. Almeida-Sousa , David Bolin , Alexandre B. Simas This is my paper · ORCID
classification math.NAcs.NAmath.STstat.TH MSC 65N3065D3260H1535R1165D3062M3060G60
keywords fractionalSPDEsGaussianrandomfieldsfiniteelementmethodsmasslumpingcovarianceapproximationnumericalquadratureMatérnmetricgraphs
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 closes a gap between the theory and practice of the SPDE approach to Gaussian random fields: the fast implementations that practitioners actually run replace the finite-element mass matrix by a diagonal 'lumped' version and evaluate integrals by quadrature, whereas existing convergence analyses covered only the idealized discretization without these shortcuts. The authors prove that these replacements are rate-preserving: for a fractional elliptic operator $L$ with spectral growth $\alpha$, the $L^2(D\times D)$ error of the mass-lumped, quadrature-based covariance approximation is bounded by the ideal Galerkin error plus a term $h^\eta$ with $\eta<\min\{4\beta-1/\alpha,2\}$. When $2\beta$ is not an integer, an additional rational-approximation step contributes only a root-exponentially decaying error in the approximation degree, and a spatially varying variance factor $\tau$ in $L^\beta(\tau u)=W$ contributes an $h^k$ error governed by the regularity of $\tau^{-1}$. This matters because it supplies explicit convergence rates for the sparse-matrix Gaussian Markov random field approximations widely used in spatial statistics and machine learning, on Euclidean domains, surfaces, and metric graphs alike.

What carries the argument

The argument rests on two second-order consistency estimates. An admissible bilinear form (Definition 3.2) satisfies $|a_L(\varphi_h,\psi_h)-a_h(\varphi_h,\psi_h)|\lesssim h^2\|\varphi_h\|_1\|\psi_h\|_1$, and an admissible inner product (Definition 3.4) satisfies $|\langle\varphi_h,\psi_h\rangle_h-(\varphi_h,\psi_h)|\lesssim h^2|\varphi_h|_1|\psi_h|_1$; the lumped-mass quadrature rule of (4.8)–(4.9) meets both bounds via Lemma 4.3. The discretized operator $L_h$ is defined by $\langle L_h\varphi,\psi\rangle_h=a_h(\varphi,\psi)$, and the map $\Lambda_h$ converts between the $L^2$ inner product and the discrete inner product, exactly accounting for the inverse mass matrices that appear in the sparse precision-matrix formulas. Fractional powers are handled through the integral representation (5.3) combined with Schatten–Hölder estimates and a spectral comparison (Proposition 5.1) showing the eigenvalues of $L_h$ are equivalent to the Galerkin eigenvalues and bounded by $h^{-2}$. A root-exponential uniform error bound for rational approximations of the scalar power function finishes the chain, making the rational degree $m$ a free parameter at logarithmic cost.

What would settle it

Compute the $L^2(S^2\times S^2)$ covariance error in the sphere experiment of Section 6.3 with $\beta=0.6$, $\kappa=1$, $\tau=1$ for the plain Galerkin approximation on meshes finer than $h\approx 0.067$ and fit the log-log slope; Theorem 4.7 predicts a rate approaching $\min\{4\beta-1,2\}=1.4$, so a slope materially below $1.4$ would falsify the surface extension of the preservation claim. Separately, run the one-dimensional experiment of Section 6.1 with a coefficient $\kappa^2$ that is Lipschitz but not twice differentiable, so $W^{2,\infty}$ regularity fails; the theory predicts the second-order consistency of Lemma 4.3 to break down and the rate to degrade below $\min\{4\beta-1/2,2\}$, so observing the full rate would show the stated assumptions are not necessary.

Watch

Extended reading notes

Core claim

The central claim is that precision-based discretizations — those built from quadrature-approximated bilinear forms and inner products — converge to the true covariance kernel at the same order as the plain Galerkin discretization. Theorem 3.5 states $\|\varrho_\beta - \varrho_h^\beta\|_{L^2(D\times D)} \lesssim \max\{\|\varrho_\beta - \hat{\varrho}_h^\beta\|_{L^2(D\times D)}, h^\eta\}$ for every $\eta<\min\{4\beta-1/\alpha,2\}$, where $\hat{\varrho}_h^\beta$ is the Galerkin covariance approximation and $\varrho_h^\beta$ the mass-lumped one. Theorem 3.7 adds to this a rational-approximation term $h^{-1/\alpha}e^{-2\pi\sqrt{\{2\beta\}m}}$ that decays root-exponentially in the rational degree $m$, so $m$ can be chosen of order $|\log h|^2$ with no effect on the spatial rate. Theorem 3.9 extends the result to models $L^\beta(\tau u)=W$ with a spatially varying variance factor: discretizing multiplication by $\tau^{-1}$ through nodal interpolation contributes a term $h^k$ when $\tau^{-1}$ belongs to the multiplier space $R^k$, so for $k\ge 2$ the variance factor does not reduce the convergence order. The results are derived in an abstract operator setting and then applied to bounded Euclidean domains, closed surfaces, and compact metric graphs, with the metric graph and sphere experiments confirming the predicted rates.

Load-bearing premise

The load-bearing premise is that the quadrature behind mass lumping is second-order consistent — approximating the true integrals to $O(h^2)$ for all trial functions, which requires $W^{2,\infty}$ coefficients and quasi-uniform meshes — and, for surfaces, the promised rate additionally rests on a Galerkin covariance estimate (Theorem 4.7) that the paper states without proof.

Editorial extensions

If this is right

  • In bounded Euclidean domains ($d\le 3$) and on compact metric graphs, the mass-lumped covariance approximation attains the same order $\min\{4\beta-d/2,2\}$ as the plain Galerkin method, so the sparse-matrix implementations of the SPDE approach are rate-preserving.
  • For non-integer $2\beta$, the rational-approximation step contributes an error that decays root-exponentially in the rational degree $m$; choosing $m$ of order $|\log h|^2$ keeps the covariance error at the spatial discretization rate.
  • For non-stationary models $L^\beta(\tau u)=W$, the discretization of the variance factor adds an $h^k$ error with $\tau^{-1}\in W^{k,\infty}$; when $k\ge 2$ the variance factor does not lower the convergence order.
  • The same preservation conclusions hold for the rational and non-stationary variants on surfaces, conditional on the stated surface Galerkin estimate (Theorem 4.7).

Reading between the lines

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

  • Because the proofs use only the two consistency estimates, the preservation result should extend to other quadrature-induced inner products beyond the standard lumped-mass rule (for example higher-order nodal quadratures), provided the $O(h^2)$ consistency holds — a testable prediction for higher-order finite elements.
  • The explicit $h^k$ dependence on $\tau$ suggests a practical rule of thumb: when the variance factor is rough ($k<2$), the accuracy of the covariance, not the fractional solver, will be the limiting factor; interpolating $\tau$ at higher order is then the cheapest way to restore the rate.
  • The theory also predicts that the often-observed 'error cancellation' in the $\beta=1$ case is only a constant-level effect: fully mass-lumped approximations can have smaller error than partially lumped ones, but both sit at the same rate — a distinction that matters for comparing implementations at fixed mesh sizes.
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, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper analyzes finite element approximations of covariance functions for fractional elliptic SPDEs driven by Gaussian white noise, in the presence of mass lumping and numerical quadrature. It introduces an abstract framework of admissible bilinear forms and admissible discrete inner products, and derives convergence estimates for precision-based covariance approximations: Theorem 3.5 shows that the precision-based approximation preserves the Galerkin covariance rate up to an h^eta term with eta < min{4beta-1/alpha,2}; Theorem 3.7 adds a root-exponentially decaying rational-approximation error; Theorem 3.9 treats the non-stationary variance-control model L^beta(tau u)=W and shows an additional h^k term controlled by the regularity of tau^{-1}. Applications are discussed for Euclidean domains, closed surfaces, and metric graphs, with numerical experiments on an interval, a metric graph, and the sphere.

Significance. If the main results are fully substantiated, the paper fills a real gap: it provides the first convergence-rate analysis of the mass-lumped, quadrature-based discretizations actually used in the SPDE approach, and it does so in a unified operator-theoretic framework that covers Euclidean domains, metric graphs, and (conditionally) surfaces. The explicit dependence of the variance-factor error on the regularity of tau is also a useful contribution. The numerical experiments in Section 6 are consistent with the stated rates and add credibility to the Euclidean and metric-graph claims. The surface part, however, is presently conditional on an unproved theorem, and the central proof relies on estimates imported from a companion paper; these points must be addressed before the advertised scope is justified.

major comments (3)
  1. [Section 4.2, Theorem 4.7 and Corollary 4.8] Theorem 4.7 is stated without proof, and the text explicitly says that the derivation is omitted and will be addressed in forthcoming work. Estimate (4.17) bounds only the difference between the precision-based and Galerkin surface covariances, not the error between the precision-based covariance and the true covariance. Without the Galerkin covariance estimate (4.18), Corollary 4.8 does not follow, and the surface/Riemannian results advertised in the abstract and Section 1.1 are unverified. Please either include a complete proof of Theorem 4.7 or clearly exclude the surface case from the paper's claims and adjust the abstract and introduction accordingly.
  2. [Section 5, proof of Theorem 3.5] The proof of the central preservation result imports several load-bearing ingredients from the companion paper [1]: the bound ||G_h|| <= C h^2, the uniform bound ||Lambda_h|| <= C, and the integral representation (5.3) from [1, Theorem 4.3]. In addition, the middle term ||bϱ_h - eϱ_h|| in the triangle inequality (5.2) is dismissed with 'follows analogously and in a simpler way' without giving the argument. Since the manuscript is not self-contained in these points, the main theorem is not fully established here. Please supply the missing estimates or provide precise and accessible statements of the imported results, and include the proof for the bϱ_h - eϱ_h term.
  3. [Section 4.1, Lemma 4.3 and Assumption 4.1] The admissibility of the quadrature bilinear form (4.9) rests on the h^2 estimates in Lemma 4.3, which are stated under W^{2,∞} coefficient regularity. Assumption 4.1 C3, however, only assumes kappa and H_{ij} belong to W^{1,∞}. For the diffusion term, the local quadrature error is ∂_i φ_h ∂_j ψ_h E_T(H); for merely Lipschitz H, E_T(H) is O(h_T^{d+1}) in d dimensions, so after multiplying by the O(h_T^{-2}) gradient product the contribution is only O(h_T^{d-1}), which is not O(h_T^2) for d >= 2. Please either strengthen Assumption 4.1 to W^{2,∞} regularity (and similarly for kappa^2 in the mass term) or prove the required h^2 consistency under the stated W^{1,∞} assumption.
minor comments (4)
  1. [Equation (4.8)] In the first expression for the lumped-mass inner product, 'phi_z(z)' should presumably read 'phi_h(z)'.
  2. [Proof of Theorem 3.5] The operator tilde{L}_h is used in the proof before being defined in the main text; please define it explicitly in Section 3, for example as the precision-based operator using the exact bilinear form a_L with the admissible inner product ⟨·,·⟩_h.
  3. [Figure 6 caption] The phrase 'with beta=0.6 and tau=1=kappa=1' is confusing; it should likely read 'tau=kappa=1'.
  4. [Section 4.1] The verification that the lumped-mass inner product satisfies Definition 3.4 uses the h^2 bound in Lemma 4.3, but Definition 3.4 is stated with the H^1 seminorm; the text should make explicit how the H^1-norm bound in (4.13) implies the seminorm bound required by the definition.

Circularity Check

1 steps flagged · score 4.0 of 10

Covariance-rate theorem is not a fitted prediction, but its key operator-comparison estimate is borrowed from the authors' own companion paper [1]; the advertised surface result additionally rests on an explicitly deferred proof (Theorem 4.7).

  1. self citation load bearing [Section 5, proof of Theorem 3.5 (around Eq. (5.3) and the estimate of E_h); Section 3.2, Definition 3.4.]
    "The lines of our proof follows similar ideas as those presented in [1, Theorem 4.2] ... From [1, Theorem 4.3], the identity ... is valid for A= eLh and A=Lh ... Reasoning as in [1, Theorem 4.2], it follows that ∥Gh∥L ≲h2 and ∥Λh∥L(L2(D),Vh) ≲1."

    The decisive operator comparison ∥Gh∥L ≲ h², which turns the h²-consistency of the lumped quadrature into closeness of the precision-based and Galerkin covariance operators, is not proved in this paper; it is taken from the authors' own companion preprint [1]. The integral representation (5.3) used in the same proof is also imported from [1, Theorem 4.3]. Theorem 3.5 therefore inherits its core spectral-comparison step from a same-author citation rather than deriving it here. This is load-bearing, although it does not make Eq. (3.8) equal to any fitted input: the final covariance statement still requires the Hilbert-Schmidt argument, the rational-approximation bound from [20], and external Galerkin rates [31], [12].

full rationale

No fitted-input-as-prediction circularity was found: Theorem 3.5, Theorem 3.7, and Theorem 3.9 are proved as inequalities from explicit admissibility hypotheses, and the numerical experiments validate, rather than define, the predicted rates. The proof machinery, however, relies substantially on the same authors' earlier work: Definitions 3.2 and 3.4 are 'borrowed from [1]', and the proof of the central estimate invokes [1, Theorem 4.3] for the Cauchy integral identity and [1, Theorem 4.2] for the key bound ∥Gh∥ ≲ h². That is a genuine self-citation load, though the cited items are technical lemmas rather than the target covariance-kernel result, so the central claim retains independent content. Separately, the surface conclusion Corollary 4.8 depends on Theorem 4.7, which is explicitly stated without proof and deferred to the authors' forthcoming work; this is an omitted-proof/completeness gap, not a circularity, and it should be counted under correctness rather than under the circularity score. The Euclidean and metric-graph Galerkin baseline rates come from independent literature ([31], [12]), which keeps the main claim externally grounded. Score 4 reflects the load-bearing self-citation while recognizing that no prediction reduces to its input by construction.

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

The abstract framework is built on standard FEM assumptions (quasi-uniform meshes, multiplier interpolation, second-order quadrature). The main risk items are the unproved surface estimate and the dependence on the authors' companion paper [1] for key spectral estimates. No fitted parameters or invented physical entities are used.

assumptions (5)
  • domain assumption Assumption 2.3: spectral Weyl-type growth of L, λ_j ≍ j^α with α>0.
    Load-bearing for trace-class conditions and all spectral summation estimates. Holds with α=2/d for Euclidean domains, α=1 for closed surfaces, α=2 for metric graphs.
  • domain assumption Assumption 3.1: existence of trial spaces V_h and interpolators I_h satisfying uniform multiplier interpolation and inverse inequalities (i)-(iv).
    Needed for admissibility of quadrature and for the h^k term in Theorem 3.9. Standard for quasi-uniform simplicial meshes with W^{k,∞} multipliers, but the surface lifted-mesh verification is asserted without full detail.
  • standard math Lemma 4.3: lumped quadrature error is second-order on each element.
    Quoted from [34, 67]; this makes the lumped inner product and lumped bilinear form admissible. If this consistency failed, the central rate-preservation result would degrade.
  • standard math Assumption 3.6 / Eq. (3.12): exists scalar approximation Q^β_m with root-exponential error; the specific bound ∥L_h^{-2β}Λ_h − R_{2β,m,h}Λ_h∥ ≲ h^{-1/α} e^{-2π√({2β}m)} is quoted from [20].
    Used in Theorem 3.7 to control the rational approximation error; the root-exponential decay is a known property of best rational and sinc approximations.
  • ad hoc to paper Theorem 4.7: surface Galerkin covariance estimate ∥ϱβ − ϱ̂^β_h∥ ≲ h^η for η<min{4β−1,2}.
    Stated without proof and deferred to 'forthcoming work' in Section 4.2; the surface application and Corollary 4.8 depend on it.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mass Lumping and Numerical Quadrature for Approximation of Fractional Elliptic Differential Equations Driven by Gaussian White Noise." pith.science (2026). https://pith.science/paper/IDGCZ3LX

@misc{pith2026260808658,
  author       = {Pith},
  title        = {Pith review of: Mass Lumping and Numerical Quadrature for Approximation of Fractional Elliptic Differential Equations Driven by Gaussian White Noise},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IDGCZ3LX}},
  note         = {Machine review of arXiv:2608.08658}
}
abstract

Fractional elliptic stochastic partial differential equations (SPDEs) are widely used in statistics and machine learning for computationally efficient and flexible modeling of Gaussian random fields. The computational efficiency of the SPDE approach relies on finite element approximations combined with numerical quadrature and mass lumping, which enable sparse matrix methods during inference. Although many works have studied finite element approximations of fractional SPDEs, the effect of the mass lumping and quadrature approximations used in practice has not been fully analyzed. To fill this gap, we derive convergence rates for numerical approximations of fractional SPDEs based on finite element discretizations combined with numerical quadrature and mass lumping. Specifically, we obtain explicit convergence rates for the mean-squared error of the covariance function in a general framework that covers the main settings where mass lumping is used in the SPDE approach. We also analyze non-stationary variance-control factors of the form $L^\beta(\tau u)=\mathcal{W}$, where $\tau$ is spatially varying, and derive covariance error estimates showing how the regularity of $\tau$ affects the convergence rate. As specific examples, we provide results for random fields on bounded Euclidean domains, Riemannian manifolds, and metric graphs. Numerical experiments are presented that confirm the theoretical results.

Figures

Figures reproduced from arXiv: 2608.08658 by the authors.

Figure 1
Figure 1. Extended computational domain and FEM mesh (top left), together with samples [PITH_FULL_IMAGE:figures/full_fig_p017_1.png] view at source ↗
Figure 2
Figure 2. Samples of Gaussian random field approximations on different approximated [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗
Figure 3
Figure 3. Sample of a Gaussian random field with precision-based covariance on the largest [PITH_FULL_IMAGE:figures/full_fig_p022_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Observed covariance errors, shown on log-log scales as functions of the mesh width [PITH_FULL_IMAGE:figures/full_fig_p028_4.png]
Figure 5
Figure 5. Figure 5: A simulation of (6.3) with β = 1/2 and κ = 1 on the metric graph used in the logo of the MetricGraph package (left). Observed L 2 (Γ × Γ) covariance errors for different values of β as functions of the mesh width h (right). The different line types correspond to theore…
Figure 6
Figure 6. Figure 6: A simulation of (6.4) with β = 0.6 and τ = 1 = κ = 1 on the unit sphere (left) and observed L 2 (S 2 × S 2 ) covariance errors for different values of β as functions of the mesh size h (right) with the same choice of τ and κ. The black dashed lines show the theoretical…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

140 extracted references · 75 canonical work pages

  1. [1]

    Oates , title =

    Emilio Porcu and Moreno Bevilacqua and Robert Schaback and Chris J. Oates , title =. Statist. Sci. , volume =. 2024 , doi =

    2024

  2. [2]

    , title =

    Stahl, Herbert R. , title =. Bull. Amer. Math. Soc. , volume =. 1993 , doi =

    1993

  3. [3]

    Harizanov, Stanislav and Lazarov, Raytcho and Margenov, Svetozar and Marinov, Pencho and Pasciak, Joseph , title =. J. Comput. Phys. , volume =. 2020 , doi =

    2020

  4. [4]

    Approximate

    Rue, H. Approximate. J. R. Stat. Soc. Ser. B Stat. Methodol. , volume =. 2009 , doi =

    2009

  5. [5]

    Lindgren, Finn and Rue, H. J. Stat. Softw. , volume =. 2015 , doi =

    2015

  6. [6]

    1999 , doi =

    Billingsley, Patrick , title =. 1999 , doi =

    1999

  7. [7]

    , title =

    Bogachev, Vladimir I. , title =. 1998 , doi =

    1998

  8. [8]

    Steinwart, Ingo , title =. J. Mach. Learn. Res. , volume =

Show all 140 references
  1. [9]

    2026 , url =

    R: A Language and Environment for Statistical Computing , organization =. 2026 , url =

    2026

  2. [10]

    2026 , note =

    Massicotte, Philippe and South, Andy , title =. 2026 , note =

    2026

  3. [11]

    Pebesma, Edzer , title =. R J. , volume =. 2018 , doi =

    2018

  4. [12]

    2026 , note =

    Lindgren, Finn , title =. 2026 , note =

    2026

  5. [13]

    , title =

    Bolin, David and Simas, Alexandre B. , title =. 2025 , note =

    2025

  6. [14]

    2014 , doi =

    Da Prato, Giuseppe and Zabczyk, Jerzy , title =. 2014 , doi =

    2014

  7. [15]

    , title =

    Stuart, Andrew M. , title =. Acta Numer. , volume =. 2010 , doi =

    2010

  8. [16]

    and Dashti, Masoumeh and Stuart, Andrew M

    Cotter, Simon L. and Dashti, Masoumeh and Stuart, Andrew M. , title =. SIAM J. Numer. Anal. , volume =. 2010 , doi =

    2010

  9. [17]

    Krainski, Elias T. and G. Advanced Spatial Modeling with Stochastic Partial Differential Equations Using. 2019 , doi =

    2019

  10. [18]

    and Steinsland, Ingelin , title =

    Ingebrigtsen, Rikke and Lindgren, Finn K. and Steinsland, Ingelin , title =. Spat. Stat. , volume =. 2014 , doi =

    2014

  11. [19]

    Exploring a new class of non-stationary spatial

    Fuglstad, Geir-Arne and Lindgren, Finn and Simpson, Daniel and Rue, H. Exploring a new class of non-stationary spatial. Statist. Sinica , volume =. 2015 , doi =

    2015

  12. [20]

    Does non-stationary spatial data always require non-stationary random fields? , journal =

    Fuglstad, Geir-Arne and Simpson, Daniel and Lindgren, Finn and Rue, H. Does non-stationary spatial data always require non-stationary random fields? , journal =. 2015 , doi =

    2015

  13. [21]

    , title =

    Bonito, Andrea and Pasciak, Joseph E. , title =. Math. Comp. , volume =. 2015 , doi =

    2015

  14. [22]

    , title =

    Bonito, Andrea and Lei, Wenyu and Pasciak, Joseph E. , title =. J. Numer. Math. , volume =. 2019 , doi =

    2019

  15. [23]

    and Wallin, Jonas , title =

    Bolin, David and Simas, Alexandre B. and Wallin, Jonas , title =. J. R. Stat. Soc. Ser. B Stat. Methodol. , year =. doi:10.1093/jrsssb/qkag074 , url =

    doi

  16. [24]

    2013 , isbn =

    Berkolaiko, Gregory and Kuchment, Peter , title =. 2013 , isbn =

    2013

  17. [25]

    and Wallin, Jonas , title =

    Bolin, David and Simas, Alexandre B. and Wallin, Jonas , title =. Bernoulli , volume =. 2024 , doi =

    2024

  18. [26]

    2026 , url =

    Raifer, Martin , title =. 2026 , url =

    2026

  19. [27]

    and Wallin, Jonas , title =

    Bolin, David and Simas, Alexandre B. and Wallin, Jonas , title =. 2025 , note =

    2025

  20. [28]

    and Wallin, Jonas , title =

    Bolin, David and Simas, Alexandre B. and Wallin, Jonas , title =. Bernoulli , volume =. 2026 , doi =

    2026

  21. [29]

    Hofreither, Clemens , title =. Comput. Math. Appl. , volume =. 2020 , doi =

    2020

  22. [30]

    Almeida-Sousa, Kelvin J. R. and Bolin, David and Simas, Alexandre B. , title =. 2026 , doi =. 2605.12082 , archivePrefix =

    2026 arXiv

  23. [31]

    Bernoulli , volume =

    Carrizo Vergara, Ricardo and Allard, Denis and Desassis, Nicolas , title =. Bernoulli , volume =. 2022 , doi =

    2022

  24. [32]

    Daniel Sanz-Alonso and Ruiyi Yang , title =. Statist. Sci. , volume =. 2022 , doi =

    2022

  25. [33]

    SIAM/ASA J

    Sanz-Alonso, Daniel and Yang, Ruiyi , title =. SIAM/ASA J. Uncertain. Quantif. , volume =. 2022 , doi =

    2022

  26. [34]

    Isotropic covariance functions on graphs and their edges , journal =

    Anderes, Ethan and M. Isotropic covariance functions on graphs and their edges , journal =. 2020 , doi =

    2020

  27. [35]

    Linear mixed effects models for non-

    Asar, \". Linear mixed effects models for non-. J. R. Stat. Soc. Ser. C Appl. Stat. , volume =. 2020 , note =

    2020

  28. [36]

    A diffusion‐based spatio‐temporal extension of

    Lindgren, Finn and Bakka, Haakon and Bolin, David and Krainski, Elias and Rue, H. A diffusion‐based spatio‐temporal extension of. Stat. Oper. Res. Trans. , volume =. 2024 , url =

    2024

  29. [37]

    and Simpson, Daniel and Rue, H

    Bakka, Haakon and Vanhatalo, Jarno and Illian, Janine B. and Simpson, Daniel and Rue, H. Non-stationary. Spat. Stat. , volume =. 2019 , doi =

    2019

  30. [38]

    2019 , doi =

    Haakon Bakka , title =. 2019 , doi =. 1803.03765 , archivePrefix =

    2019 arXiv

  31. [39]

    2020 , url =

    Bakka, Haakon and Krainski, Elias and Bolin, David and Rue, Håvard and Lindgren, Finn , title =. 2020 , url =. doi:10.48550/arXiv.2006.04917 , eprint =

    2020 doi arXiv

  32. [40]

    Biometrika , volume =

    Besag, Julian and Kooperberg, Charles , title =. Biometrika , volume =. 1995 , note =

    1995

  33. [41]

    Biometrika , volume =

    Besag, Julian and Mondal, Debashis , title =. Biometrika , volume =. 2005 , note =

    2005

  34. [42]

    Besag, Julian , title =. J. R. Stat. Soc. Ser. B , volume =. 1974 , url =

    1974

  35. [43]

    and Weiss, D

    Bhatt, S. and Weiss, D. J. and Cameron, E. and Bisanzio, D. and Mappin, B. and Dalrymple, U. and Battle, K. E. and Moyes, C. L. and Henry, A. and Eckhoff, P. A. and Wenger, E. A. and Briët, O. and Penny, M. A. and Smith, T. A. and Bennett, A. and Yukich, J. and Eisele, T. P. a...

    2015

  36. [44]

    Bernoulli , volume =

    Bolin, David and Kirchner, Kristin , title =. Bernoulli , volume =. 2023 , doi =

    2023

  37. [45]

    Bolin, David and Kirchner, Kristin , title =. J. Comput. Graph. Statist. , volume =. 2020 , doi =

    2020

  38. [46]

    Weak convergence of

    Bolin, David and Kirchner, Kristin and Kov\'. Weak convergence of. BIT Numer. Math. , volume =. 2018 , doi =

    2018

  39. [47]

    Numerical solution of fractional elliptic stochastic

    Bolin, David and Kirchner, Kristin and Kov\'. Numerical solution of fractional elliptic stochastic. IMA J. Numer. Anal. , volume =. 2020 , doi =

    2020

  40. [48]

    Bolin, David and Lindgren, Finn , title =. Ann. Appl. Stat. , volume =. 2011 , doi =

    2011

  41. [49]

    Bolin, David , title =. Scand. J. Statist. , volume =. 2014 , url =

    2014

  42. [50]

    and Xiong, Zhen , title =

    Bolin, David and Simas, Alexandre B. and Xiong, Zhen , title =. J. Comput. Graph. Statist. , volume =. 2024 , doi =

    2024

  43. [51]

    Regularity and numerical approximation of fractional elliptic differential equations on compact metric graphs , journal =

    Bolin, David and Kov. Regularity and numerical approximation of fractional elliptic differential equations on compact metric graphs , journal =. 2024 , doi =

    2024

  44. [52]

    Bolin, David and Lindgren, Finn , title =. Comput. Statist. Data Anal. , volume =. 2013 , doi =

    2013

  45. [53]

    Pasciak , title =

    Andrea Bonito and Joseph E. Pasciak , title =. Math. Comp. , volume =. 2015 , doi =

    2015

  46. [54]

    Bonito, Andrea and Guignard, Diane and Lei, Wenyu , title =. Comput. Methods Appl. Math. , volume =. 2024 , doi =

    2024

  47. [55]

    Borovitskiy, Viacheslav and Terenin, Alexander and Mostowsky, Peter and Deisenroth, Marc Peter , title =

  48. [56]

    2020 , doi =

    Borovitskiy, Viacheslav and Terenin, Alexander and Mostowsky, Peter and Deisenroth, Marc Peter , title =. 2020 , doi =. 2006.10160 , archivePrefix =

    2020 arXiv

  49. [57]

    and Scott, L

    Brenner, Susanne C. and Scott, L. Ridgway , title =. 2008 , doi =

    2008

  50. [58]

    Bolin, David and Wallin, Jonas , title =. J. R. Stat. Soc. Ser. B Stat. Methodol. , volume =. 2020 , doi =

    2020

  51. [59]

    AStA Adv

    Cameletti, Michela and Lindgren, Finn and Simpson, Daniel and Rue, Håvard , title =. AStA Adv. Stat. Anal. , volume =. 2013 , url =

    2013

  52. [60]

    and Roininen, Lassi , title =

    Suuronen, Jarkko and Chada, Neil K. and Roininen, Lassi , title =. Stat. Comput. , volume =. 2022 , doi =

    2022

  53. [61]

    Error estimates for a finite volume element method for parabolic equations in convex polygonal domains , journal =

    Chatzipantelidis, P and Lazarov, RD and Thom. Error estimates for a finite volume element method for parabolic equations in convex polygonal domains , journal =. 2004 , doi =

    2004

  54. [62]

    , title =

    Ciarlet, Philippe G. , title =. 2002 , doi =

    2002

  55. [63]

    and Wilkinson, Richard D

    Coveney, Sam and Corrado, Cesare and Roney, Caroline H. and Wilkinson, Richard D. and Oakley, Jeremy E. and Lindgren, Finn and Williams, Steven E. and O’Neill, Mark D. and Niederer, Steven A. and Clayton, Richard H. , title =. IEEE Trans. Biomed. Eng. , volume =. 2020 , note =

    2020

  56. [64]

    and Kirchner, Kristin , title =

    Cox, Sonja G. and Kirchner, Kristin , title =. Numer. Math. , volume =. 2020 , doi =

    2020

  57. [65]

    Wikle , title =

    Noel Cressie and Christopher K. Wikle , title =. 2011 , doi =

    2011

  58. [66]

    and Gelfand, Alan E

    Datta, Abhirup and Banerjee, Sudipto and Finley, Andrew O. and Gelfand, Alan E. , title =. J. Amer. Statist. Assoc. , volume =. 2016 , note =

    2016

  59. [67]

    Davies, E. B. , title =. 1995 , doi =

    1995

  60. [68]

    Surface Finite Element Approximation of Spherical

    Jansson, Erik and Kov\'. Surface Finite Element Approximation of Spherical. SIAM J. Sci. Comput. , volume =. 2022 , url =

    2022

  61. [69]

    and Wu, Hau-Tieng and Wu, Nan , title =

    Dunson, David B. and Wu, Hau-Tieng and Wu, Nan , title =. J. R. Stat. Soc. Ser. B Stat. Methodol. , volume =. 2022 , url =. doi:10.1111/rssb.12486 , eprint =

    2022 doi

  62. [70]

    , title =

    Dziuk, Gerhard and Elliott, Charles M. , title =. Acta Numer. , volume =. 2013 , doi =

    2013

  63. [71]

    2018 , doi =

    Evans, LawrenceCraig , title =. 2018 , doi =

    2018

  64. [72]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages =

    Feragen, Aasa and Lauze, Francois and Hauberg, Soren , title =. Proceedings of the IEEE conference on computer vision and pattern recognition , pages =. 2015 , doi =

    2015

  65. [73]

    Interpolating between optimal transport and mmd using sinkhorn divergences , booktitle =

    Feydy, Jean and S. Interpolating between optimal transport and mmd using sinkhorn divergences , booktitle =. 2019 , doi =

    2019

  66. [74]

    , title =

    Fix, George J. , title =. The mathematical foundations of the finite element method with applications to partial differential equations , pages =. 1972 , doi =

    1972

  67. [75]

    Fuglstad, Geir-Arne and Simpson, Daniel and Lindgren, Finn and Rue, Håvard , title =. Spat. Stat. , volume =. 2015 , doi =

    2015

  68. [76]

    Gneiting, Tilmann and Raftery, Adrian E , title =. J. Amer. Statist. Assoc. , volume =. 2007 , note =

    2007

  69. [77]

    2011 , doi =

    Grisvard, Pierre , title =. 2011 , doi =

    2011

  70. [78]

    Guella, Jean Carlo , title =. J. Approx. Theory , volume =. 2022 , doi =

    2022

  71. [79]

    Harbrecht, Helmut and Herrmann, Lukas and Kirchner, Kristin and Schwab, Christoph , title =. Adv. Comput. Math. , volume =. 2024 , doi =

    2024

  72. [80]

    Herrmann, Lukas and Kirchner, Kristin and Schwab, Christoph , title =. Math. Models Methods Appl. Sci. , volume =. 2020 , doi =

    2020

  73. [81]

    Hildeman, Anders and Bolin, David and Rychlik, Igor , title =. Spat. Stat. , volume =. 2021 , doi =

    2021

  74. [82]

    Ingebrigtsen, Rikke and Lindgren, Finn and Steinsland, Ingelin , title =. Spat. Stat. , series =. 2014 , doi =

    2014

  75. [83]

    Jin, Bangti and Lazarov, Raytcho and Zhou, Zhi , title =. SIAM J. Numer. Anal. , volume =. 2013 , doi =

    2013

  76. [84]

    Katzfuss, Matthias , title =. J. Amer. Statist. Assoc. , volume =. 2017 , note =

    2017

  77. [85]

    Kirchner, Kristin and Bolin, David , title =. Ann. Statist. , volume =. 2022 , doi =

    2022

  78. [86]

    Lang, Annika and Schwab, Christoph , title =. Ann. Appl. Probab. , volume =. 2015 , url =. doi:10.1214/14-AAP1067 , mrreviewer =

    2015 doi

  79. [87]

    BIT Numer

    Liang, Shengde and Ma, Xiuling and Zhou, Aihui , title =. BIT Numer. Math. , volume =. 2001 , doi =

    2001

  80. [88]

    and Dean, Joseph P

    Baratta, Igor A. and Dean, Joseph P. and Dokken, J. 2023 , doi =

    2023

  81. [89]

    Internat

    Geuzaine, Christophe and Remacle, Jean-Fran. Internat. J. Numer. Methods Engrg. , volume =. 2009 , doi =

    2009

  82. [90]

    Unified Form Language: A domain-specific language for weak formulations of partial differential equations , journal =

    Aln. Unified Form Language: A domain-specific language for weak formulations of partial differential equations , journal =. 2014 , doi =

    2014

  83. [91]

    and Paz, Rodrigo R

    Dalcin, Lisandro D. and Paz, Rodrigo R. and Kler, Pablo A. and Cosimo, Alejandro , title =. Adv. Water Resour. , volume =. 2011 , doi =

    2011

  84. [92]

    2018 , url =

    Gallichan, Daniel , title =. 2018 , url =

    2018

  85. [93]

    An explicit link between

    Lindgren, Finn and Rue, H. An explicit link between. J. R. Stat. Soc. Ser. B Stat. Methodol. , volume =. 2011 , doi =

    2011

  86. [94]

    Lindgren, Finn and Bolin, David and Rue, H. The. Spat. Stat. , volume =. 2022 , doi =

    2022

  87. [95]

    Liu, Xiaoyu and Guillas, Serge and Lai, Ming-Jun , title =. J. Comput. Graph. Statist. , volume =. 2016 , url =

    2016

  88. [96]

    Spatial variation:

    Mat\'. Spatial variation:. 1960 , doi =

    1960

  89. [97]

    and Yue, Yu and Bolin, David and Lindgren, Finn and Lindquist, Martin A

    Mejia, Amanda F. and Yue, Yu and Bolin, David and Lindgren, Finn and Lindquist, Martin A. , title =. J. Amer. Statist. Assoc. , volume =. 2020 , url =

    2020

  90. [98]

    Mejia, Amanda F and Bolin, David and Yue, Yu Ryan and Wang, Jiongran and Caffo, Brian S and Nebel, Mary Beth , title =. J. Comput. Graph. Statist. , volume =. 2023 , doi =

    2023

  91. [99]

    Miller and Eric Rexstad and Len Thomas and Laura Marshall and Jeffrey L

    David L. Miller and Eric Rexstad and Len Thomas and Laura Marshall and Jeffrey L. Laake , title =. J. Stat. Softw. , volume =. 2019 , doi =

    2019

  92. [100]

    Miller and Eric Rexstad and Louise Burt and Mark V

    David L. Miller and Eric Rexstad and Louise Burt and Mark V. Bravington and Sharon Hedley. , title =. 2021 , note =

    2021

  93. [101]

    and Glennie, Richard and Seaton, Andrew E

    Miller, David L. and Glennie, Richard and Seaton, Andrew E. , title =. J. Agric. Biol. Environ. Stat. , volume =. 2020 , url =

    2020

  94. [102]

    1959 , doi =

    Parzen, Emanuel , title =. 1959 , doi =

    1959

  95. [103]

    , title =

    Peruzzi, Michele and Banerjee, Sudipto and Finley, Andrew O. , title =. J. Amer. Statist. Assoc. , volume =. 2022 , url =

    2022

  96. [104]

    Minh, Ha Quang , title =. Anal. Appl. (Singap.) , volume =. 2023 , url =

    2023

  97. [105]

    and Auchmann, Renate and Bessembinder, Janette and Brönnimann, Stefan and Brugnara, Yuri and Capponi, Francesco and Carrea, Laura and Dodd, Emma M

    Rayner, Nick A. and Auchmann, Renate and Bessembinder, Janette and Brönnimann, Stefan and Brugnara, Yuri and Capponi, Francesco and Carrea, Laura and Dodd, Emma M. A. and Ghent, Darren and Good, Elizabeth and Høyer, Jacob L. and Kennedy, John J. and Kent, Elizabeth C. and Kill...

    2020

  98. [106]

    Roininen, Lassi and Huttunen, Janne M. J. and Lasanen, Sari , title =. Inverse Probl. Imaging , volume =. 2014 , url =

    2014

  99. [107]

    Inverse Probl

    Roininen, Lassi and Girolami, Mark and Lasanen, Sari and Markkanen, Markku , title =. Inverse Probl. Imaging , volume =. 2019 , url =

    2019

  100. [108]

    Rozanov, Ju A , title =. Math. USSR-Sb. , volume =. 1977 , doi =

    1977

  101. [109]

    and Xiong, Zhen , title =

    Bolin, David and Simas, Alexandre B. and Xiong, Zhen , title =. 2021 , note =

    2021

  102. [110]

    and Held, L

    Rue, H. and Held, L. , title =. 2005 , url =. doi:10.1201/9780203492024 , added-at =

    2005 doi

  103. [111]

    and Guttorp, Peter , title =

    Sampson, Paul D. and Guttorp, Peter , title =. J. Amer. Statist. Assoc. , volume =. 1992 , note =

    1992

  104. [112]

    and Ramsay, James O

    Sangalli, Laura M. and Ramsay, James O. and Ramsay, Timothy O. , title =. J. R. Stat. Soc. Ser. B Stat. Methodol. , volume =. 2013 , note =

    2013

  105. [113]

    , title =

    Sangalli, Laura M. , title =. Internat. Statist. Rev. , volume =. 2021 , note =

    2021

  106. [114]

    Bayesian Anal

    Sid. Bayesian Anal. , volume =. 2021 , url =

    2021

  107. [115]

    and Illian, J

    Simpson, D. and Illian, J. B. and Lindgren, F. and S rbye, S. H. and Rue, H. , title =. Biometrika , volume =. 2016 , url =

    2016

  108. [116]

    Environmetrics , volume =

    Simpson, Daniel and Lindgren, Finn and Rue, Håvard , title =. Environmetrics , volume =. 2012 , note =

    2012

  109. [117]

    , title =

    Spencer, Daniel and Yue, Yu Ryan and Bolin, David and Ryan, Sarah and Mejia, Amanda F. , title =. NeuroImage , volume =. 2022 , doi =

    2022

  110. [118]

    Stein, Michael L , title =. J. Amer. Statist. Assoc. , volume =. 2005 , doi =

    2005

  111. [119]

    , title =

    Stein, Michael L. , title =. 1999 , doi =

    1999

  112. [120]

    Galerkin finite element methods for parabolic problems , series =

    Thom. Galerkin finite element methods for parabolic problems , series =. 2007 , doi =

    2007

  113. [121]

    Chou, So-Hsiang and Li, Qian , title =. Math. Comp. , volume =. 2000 , doi =

    2000

  114. [122]

    Tsukuda, K\^oji , title =

  115. [123]

    and Scarabosio, L

    Khristenko, U. and Scarabosio, L. and Swierczynski, P. and Ullmann, E. and Wohlmuth, B. , title =. SIAM/ASA J. Uncertain. Quantif. , volume =. 2019 , url =

    2019

  116. [124]

    and Wellner, Jon A

    van der Vaart, Aad W. and Wellner, Jon A. , title =. 1996 , doi =

    1996

  117. [125]

    Vehtari, Aki and Gelman, Andrew and Gabry, Jonah , title =. Stat. Comput. , volume =. 2017 , url =

    2017

  118. [126]

    , title =

    Walder, Adam and Hanks, Ephraim M. , title =. Comput. Statist. Data Anal. , volume =. 2020 , url =

    2020

  119. [127]

    Wallin, Jonas and Bolin, David , title =. Scand. J. Statist. , volume =. 2015 , url =

    2015

  120. [128]

    2021 , publisher =

    Olav Kallenberg , title =. 2021 , publisher =

    2021

  121. [129]

    Biometrika , volume =

    Whittle, Peter , title =. Biometrika , volume =. 1954 , url =

    1954

  122. [130]

    , title =

    Whittle, P. , title =. Bull. Inst. Internat. Statist. , volume =

  123. [131]

    Wikle, C. K. and Zammit-Mangion, A. and Cressie, N. , title =. 2019 , doi =

    2019

  124. [132]

    Zhang, Hao , title =. J. Amer. Statist. Assoc. , volume =. 2004 , doi =

    2004

  125. [133]

    Zhang, Ran and Czado, Claudia and Sigloch, Karin , title =. J. R. Stat. Soc. Ser. C Appl. Stat. , volume =. 2016 , note =

    2016

  126. [134]

    2012 , doi =

    Zeidler, Eberhard , title =. 2012 , doi =

    2012

  127. [135]

    1995 , doi =

    Nualart, David , title =. 1995 , doi =

    1995

  128. [136]

    , title =

    Lee, John M. , title =. 2013 , doi =

    2013

  129. [137]

    and Elliott, C

    Dziuk, G. and Elliott, C. M. , title =. IMA J. Numer. Anal. , volume =. 2007 , url =

    2007

  130. [138]

    , title =

    Cohn, Donald L. , title =. 2013 , doi =

    2013

  131. [139]

    2010 , doi =

    Simon, Barry , title =. 2010 , doi =

    2010

  132. [140]

    , title =

    Bolin, David and Saduakhas, Damilya and Simas, Alexandre B. , title =. Biometrika , year =. doi:10.1093/biomet/asag043 , url =

    doi

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.