pith. sign in

arxiv: 2505.13929 · v5 · submitted 2025-05-20 · 🧮 math.NA · cs.NA

Error estimates for numerical approximations of a nonlinear gradient flow model

Jerome Droniou , Kim-Ngan Le , Huateng Zhu This is my paper

Pith reviewed 2026-05-22 15:09 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords nonlinear gradient flowgradient discretisation methoderror estimatesimplicit schemefinite element methodsminimal surfacetotal variation flownumerical analysis
0
0 comments X

The pith

A fully discrete implicit scheme for a nonlinear gradient flow achieves provable error estimates via the gradient discretisation method.

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

The paper applies the gradient discretisation method to a nonlinear gradient flow that can be viewed as a parabolic minimal surface problem or a regularised total variation flow. It introduces a fully discretised implicit scheme, proves existence and uniqueness of its solutions, and analyses the scheme's stability and consistency. These properties are then used to derive error estimates between the numerical approximations and the exact solution of the continuous model. The analysis covers a broad class of discretisation techniques, including conforming and nonconforming finite elements.

Core claim

Using the gradient discretisation method, a fully discretised implicit scheme for the nonlinear gradient flow is constructed for which existence and uniqueness of solutions are established, along with stability, consistency, and error estimates.

What carries the argument

The gradient discretisation method (GDM), a unified convergence analysis framework covering conforming and nonconforming numerical methods such as finite elements and two-point flux approximations.

If this is right

  • The implicit scheme admits a unique solution at each time step.
  • Stability estimates bound the discrete solution independently of the mesh size and time step.
  • The scheme is consistent with the continuous nonlinear gradient flow problem.
  • Error estimates hold in suitable norms for the difference between numerical and exact solutions.

Where Pith is reading between the lines

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

  • The same analysis approach could be applied to other regularised evolution equations with similar nonlinear structure.
  • Practical simulations of minimal surface flow might use the error bounds to decide when to stop mesh refinement.
  • The framework could be tested against alternative time-stepping methods to compare observed versus theoretical accuracy.

Load-bearing premise

The gradient discretisation method framework extends to this nonlinear parabolic minimal surface or regularised total variation model while preserving the consistency and stability properties required for the error analysis.

What would settle it

Numerical experiments on successively refined meshes that fail to exhibit the predicted convergence rates in the error estimates between discrete and continuous solutions.

Figures

Figures reproduced from arXiv: 2505.13929 by Huateng Zhu, Jerome Droniou, Kim-Ngan Le.

Figure 1
Figure 1. Figure 1: Test 1: Errors w.r.t. h: CP1FEM vs. NCP1FEM [PITH_FULL_IMAGE:figures/full_fig_p024_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Test 2: Errors w.r.t. h of CP1FEM with ρ = 1, h1/2 , h. 10−3 10−2 10−1 100 10−4 10−3 10−2 10−1 100 101 102 ρ mesh4 mesh5 mesh6 slope −1 (a) Error E (ρ) 1 10−3 10−2 10−1 100 10−3 10−2 10−1 100 101 102 ρ mesh4 mesh5 mesh6 (b) Error E (ρ) 2 10−3 10−2 10−1 100 10−3 10−2 10−1 100 101 102 ρ mesh4 mesh5 mesh6 (c) Error E (ρ) 3 [PITH_FULL_IMAGE:figures/full_fig_p025_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Test 2: Errors w.r.t. ρ of CP1FEM on mesh4, mesh5, and mesh6. while the errors E2 and E3 of CP1FEM (see Figures 2 and 3) are slightly lower than those of NCP1FEM (see Figures 4 and 5). A key observation related to the parameter ρ is its impact on the error magnitude and the convergence rate. As ρ decreases, the magnitude of E (ρ) 1 increases for both schemes, yet the magnitudes of E (ρ) 2 and E (ρ) 3 remai… view at source ↗
Figure 4
Figure 4. Figure 4: Test 2: Errors w.r.t. h of NCP1FEM with ρ = 1, h1/2 , h. 10−3 10−2 10−1 100 10−4 10−3 10−2 10−1 100 101 102 ρ mesh4 mesh5 mesh6 slope −1 (a) Error E (ρ) 1 10−3 10−2 10−1 100 10−3 10−2 10−1 100 101 102 ρ mesh4 mesh5 mesh6 (b) Error E (ρ) 2 10−3 10−2 10−1 100 10−3 10−2 10−1 100 101 102 ρ mesh4 mesh5 mesh6 (c) Error E (ρ) 3 [PITH_FULL_IMAGE:figures/full_fig_p026_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Test 2: Errors w.r.t. ρ of NCP1FEM on mesh4, mesh5, and mesh6. which is imposed with a non-smooth initial condition u0 = S(x, y) + x(1 − x)y(1 − y) + 0.251B, (x, y) ∈ Ω. where S(x, y) = 0.25 + ln  cos(y − 0.5) cos(x − 0.5) and B := {(x, y) : |x − 0.5| 2 + |y − 0.5| 2 ⩽ 0.16}; and boundary conditions u|∂Ω = S|∂Ω. Since Ω is convex and the surface boundary S|∂Ω admits a bijective projection with ∂Ω, there … view at source ↗
Figure 6
Figure 6. Figure 6: Test 3: CP1FEM, snapshots of numerical solution at the time t. (a) t = 0 (b) t = 0.1 (c) t = 1 [PITH_FULL_IMAGE:figures/full_fig_p027_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Test 3: NCP1FEM, snapshots of numerical solution at the time t. (a) CP1FEM (b) NCP1FEM [PITH_FULL_IMAGE:figures/full_fig_p027_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Test 3: comparison of numerical solution at the time T and the surface S (in red wireframe). 7. Conclusion We established error estimates for the gradient scheme, in R 2 , which holds true for all the conforming and nonconforming approximations that are in the framework of the gradient dis￾cretisation method. Moreover, the minimal surface-like interpolation and its relevant error estimates we proved in thi… view at source ↗
read the original abstract

We perform numerical analysis of a nonlinear gradient flow, which can be regarded as a parabolic minimal surface problem or a regularised total variation flow, using the gradient discretisation method (GDM). GDM is a unified convergence analysis framework that covers conforming and nonconforming numerical methods, for instance, conforming and nonconforming finite element, two-point flux approximation, etc.. In this paper, a fully discretised implicit scheme of the model is proposed, the existence and uniqueness of the solution to the scheme is proved, the stability and consistency of the scheme are analysed, and error estimates are established. Numerical results based on the conforming and nonconforming $\mathbb{P}^1$ finite elements are also provided.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript applies the gradient discretisation method (GDM) to a nonlinear gradient flow that models a parabolic minimal-surface problem or regularised total-variation flow. It introduces a fully discrete implicit scheme, proves existence and uniqueness of the discrete solution, establishes stability and consistency of the scheme, derives error estimates, and reports numerical experiments using conforming and nonconforming P1 finite elements.

Significance. A rigorous extension of GDM error analysis to this nonlinear parabolic setting would supply a unified convergence theory covering multiple discretisation families for a problem class arising in geometry and image processing; the provision of both theoretical estimates and concrete numerical illustrations strengthens the contribution if the regularity requirements are made explicit.

major comments (2)
  1. [§4 and §5] §4 (stability) and §5 (error analysis): the L^∞(0,T;H¹) bound obtained from the implicit scheme is insufficient by itself to close the consistency estimate for the nonlinear flux F(ξ)=ξ/√(1+|ξ|²). The difference |F(∇u)-F(∇_D u_D)| requires either a uniform Lipschitz bound on F or control of second derivatives of the exact solution; neither is supplied by the preceding stability result nor stated as an additional hypothesis on u.
  2. [Theorem 5.1] Theorem 5.1 (error estimate): the stated convergence rate appears to rest on an implicit W^{2,p} or ∇u∈L^∞ assumption that is not justified by the weak stability result alone; without this, the consistency term for the nonlinear operator cannot be bounded at the claimed order.
minor comments (2)
  1. [Abstract] The abstract states that numerical results are provided but does not indicate the specific test problems, mesh sizes, or observed convergence rates.
  2. [§5] Notation for the discrete gradient operator ∇_D and the reconstruction operators should be recalled at the beginning of the error-analysis section for readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We address the major comments point by point below, indicating the revisions that will be incorporated to make the regularity assumptions explicit and close the estimates rigorously.

read point-by-point responses

  1. Referee: [§4 and §5] §4 (stability) and §5 (error analysis): the L^∞(0,T;H¹) bound obtained from the implicit scheme is insufficient by itself to close the consistency estimate for the nonlinear flux F(ξ)=ξ/√(1+|ξ|²). The difference |F(∇u)-F(∇_D u_D)| requires either a uniform Lipschitz bound on F or control of second derivatives of the exact solution; neither is supplied by the preceding stability result nor stated as an additional hypothesis on u.

    Authors: We agree that the L^∞(0,T; H¹) stability bound alone does not automatically yield a uniform Lipschitz constant for F without further control. In the revised manuscript we will add an explicit hypothesis that the exact solution satisfies ∇u ∈ L^∞(Ω × (0,T)). Under this assumption F is Lipschitz on the relevant bounded range, which permits a direct estimate of the consistency term |F(∇u) − F(∇_D u_D)| at the order required by the error analysis. The hypothesis will be stated at the beginning of Section 5 and referenced in the proof of Theorem 5.1. revision: yes

  2. Referee: [Theorem 5.1] Theorem 5.1 (error estimate): the stated convergence rate appears to rest on an implicit W^{2,p} or ∇u∈L^∞ assumption that is not justified by the weak stability result alone; without this, the consistency term for the nonlinear operator cannot be bounded at the claimed order.

    Authors: The rate claimed in Theorem 5.1 does rely on sufficient regularity of u to control the nonlinear consistency error. While the discrete stability result supplies an L^∞(0,T; H¹) bound, the continuous solution must satisfy an additional regularity assumption (∇u ∈ L^∞ or an equivalent W^{2,p} condition) for the consistency term to be bounded at the stated order. We will revise the statement of Theorem 5.1 to list this regularity hypothesis explicitly and will add a short remark explaining why it is needed for the nonlinear flux. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation applies standard GDM to new model

full rationale

The paper proposes a fully discrete implicit scheme for the nonlinear gradient flow, proves existence and uniqueness directly for the scheme, establishes stability via discrete energy estimates, and derives consistency and error estimates by applying the general GDM consistency and stability properties to the specific nonlinearity F(ξ) = ξ / sqrt(1 + |ξ|^2). No step equates a 'prediction' to a fitted input by construction, redefines terms self-referentially, or reduces the central error bound to a self-citation chain whose assumptions already embed the target result. GDM is invoked as an external unified framework whose prior proofs are independent of this model's error rates; the application to the regularised minimal-surface flow supplies new content (scheme-specific monotonicity arguments and discrete estimates) without tautological reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The analysis depends on standard well-posedness assumptions for the continuous nonlinear gradient flow and on the abstract properties of the gradient discretisation method; no free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption The nonlinear gradient flow model admits a sufficiently regular weak solution in appropriate function spaces.
    Required for the consistency and error estimate statements to be meaningful.

pith-pipeline@v0.9.0 · 5644 in / 1185 out tokens · 47527 ms · 2026-05-22T15:09:30.295387+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

46 extracted references · 46 canonical work pages

  1. [1]

    Error estimates for a finite element approximation of a minimal surface

    Claes Johnson and Vidar Thomée. “Error estimates for a finite element approximation of a minimal surface”. In:Mathematics of Computation29.130 (1975), pp. 343–349. 27

    work page 1975

  2. [2]

    Some asymptotic error estimates for finite element approximation of minimal surfaces

    Rolf Rannacher. “Some asymptotic error estimates for finite element approximation of minimal surfaces”. In:RAIRO. Analyse numérique11.2 (1977), pp. 181–196

    work page 1977

  3. [3]

    Pseudosolutions of the time-dependent minimal surface problem

    Alain Lichnewsky and Roger Temam. “Pseudosolutions of the time-dependent minimal surface problem”. In:Journal of differential equations30.3 (1978), pp. 340–364

    work page 1978

  4. [4]

    Evolutionary surfaces of prescribed mean curvature

    Claus Gerhardt. “Evolutionary surfaces of prescribed mean curvature”. In:Journal of Differential Equations36.1 (1980), pp. 139–172

    work page 1980

  5. [5]

    Nonlinear total variation based noise removal algorithms

    Leonid I. Rudin, S. Osher, and Emad Fatemi. “Nonlinear total variation based noise removal algorithms”. In:Physica D: Nonlinear Phenomena60 (1992), pp. 259–268

    work page 1992

  6. [6]

    Convergence of a finite element method for non- parametric mean curvature flow

    Klaus Deckelnick and Gerhard Dziuk. “Convergence of a finite element method for non- parametric mean curvature flow”. In:Numerische Mathematik72.2 (1995), pp. 197–222

    work page 1995

  7. [7]

    Image recovery via total variation minimiza- tion and related problems

    Antonin Chambolle and Pierre-Louis Lions. “Image recovery via total variation minimiza- tion and related problems”. In:Numerische Mathematik76 (1997), pp. 167–188

    work page 1997

  8. [8]

    Numerical solution of para- bolic equations related to level set formulation of mean curvature flow

    Angela Handlovičová, Karol Mikula, and Alessandro Sarti. “Numerical solution of para- bolic equations related to level set formulation of mean curvature flow”. In:Computing and visualization in Science1.3 (1998), pp. 179–182

    work page 1998

  9. [9]

    Equations with singular diffusivity

    R Kobayashi and Yt Giga. “Equations with singular diffusivity”. In:Journal of statistical physics95 (1999), pp. 1187–1220

    work page 1999

  10. [10]

    Minimizingtotalvariationflow

    FuensantaAndreuetal.“Minimizingtotalvariationflow”.In:Comptes Rendus de l’Académie des Sciences-Series I-Mathematics331.11 (2000), pp. 867–872

    work page 2000

  11. [11]

    Image denoising and segmentation via non- linear diffusion

    Yunmei Chen, Baba C Vemuri, and Li Wang. “Image denoising and segmentation via non- linear diffusion”. In:Computers & Mathematics with Applications39.5-6 (2000), pp. 131– 149

    work page 2000

  12. [12]

    Error estimates for a semi-implicit fully discrete finite element scheme for the mean curvature flow of graphs

    Klaus Deckelnick and Gerhard Dziuk. “Error estimates for a semi-implicit fully discrete finite element scheme for the mean curvature flow of graphs”. In:Interfaces and Free Boundaries2.4 (2000), pp. 341–359

    work page 2000

  13. [13]

    The Dirichlet problem for the total variation flow

    Fuensanta Andreu et al. “The Dirichlet problem for the total variation flow”. In:Journal of Functional Analysis180.2 (2001), pp. 347–403

    work page 2001

  14. [14]

    Some qualitative properties for the total variation flow

    Fuensanta Andreu et al. “Some qualitative properties for the total variation flow”. In: Journal of functional analysis188.2 (2002), pp. 516–547

    work page 2002

  15. [15]

    The total variation flow in RN

    Giovanni Bellettini, Vicent Caselles, and Matteo Novaga. “The total variation flow in RN”. In:Journal of Differential Equations184.2 (2002), pp. 475–525

    work page 2002

  16. [16]

    On the role of the BV image model in image restora- tion

    Tony F Chan and Jianhong Shen. “On the role of the BV image model in image restora- tion”. In:Contemporary Mathematics330 (2003), pp. 25–42

    work page 2003

  17. [17]

    Analysis of total variation flow and its finite element approximations

    Xiaobing H. Feng and Andreas Prohl. “Analysis of total variation flow and its finite element approximations”. In:Mathematical Modelling and Numerical Analysis37 (2003), pp. 533–556

    work page 2003

  18. [18]

    Rate of convergence of regu- larization procedures and finite element approximations for the total variation flow

    Xiaobing Feng, Markus von Oehsen, and Andreas Prohl. “Rate of convergence of regu- larization procedures and finite element approximations for the total variation flow”. In: Numerische Mathematik100 (2005), pp. 441–456

    work page 2005

  19. [19]

    Error analysis of a finite element method for the Willmore flow of graphs

    Klaus Deckelnick and Gerhard Dziuk. “Error analysis of a finite element method for the Willmore flow of graphs”. In:Interfaces and free boundaries8.1 (2006), pp. 21–46

    work page 2006

  20. [20]

    On p-Harmonic Map Heat Flows for 1⩽p⩽∞and Their Finite Element Approximations

    John W Barrett, Xiaobing Feng, and Andreas Prohl. “On p-Harmonic Map Heat Flows for 1⩽p⩽∞and Their Finite Element Approximations”. In:SIAM journal on mathematical analysis40.4 (2008), pp. 1471–1498

    work page 2008

  21. [21]

    TVSeg-Interactive Total Variation Based Image Segmentation

    Markus Unger et al. “TVSeg-Interactive Total Variation Based Image Segmentation.” In: BMVC. Vol. 31. Citeseer. 2008, pp. 44–46

    work page 2008

  22. [22]

    A new alternating minimization algorithm for total variation image reconstruction

    Yilun Wang et al. “A new alternating minimization algorithm for total variation image reconstruction”. In:SIAM Journal on Imaging Sciences1.3 (2008), pp. 248–272

    work page 2008

  23. [23]

    Finite element approximation of the gradient flow for a classoflineargrowthenergieswithapplicationstocolorimagedenoising

    Xiaobing Feng and Miun Yoon. “Finite element approximation of the gradient flow for a classoflineargrowthenergieswithapplicationstocolorimagedenoising”.In:International Journal of Numerical Analysis & Modeling6.3 (2009). 28

    work page 2009

  24. [24]

    Image sharpening via Sobolev gradient flows

    Jeff Calder, A Mansouri, and Anthony Yezzi. “Image sharpening via Sobolev gradient flows”. In:SIAM Journal on Imaging Sciences3.4 (2010), pp. 981–1014

    work page 2010

  25. [25]

    Minimalsurfaces

    UlrichDierkesetal.“Minimalsurfaces”.In:Minimal Surfaces I: Boundary Value Problems. Springer, 2010, pp. 53–88

    work page 2010

  26. [26]

    Very singular diffusion equations: second and fourth or- der problems

    Mi-Ho Giga and Yoshikazu Giga. “Very singular diffusion equations: second and fourth or- der problems”. In:Japan journal of industrial and applied mathematics27 (2010), pp. 323– 345

    work page 2010

  27. [27]

    Total variation minimization with finite elements: convergence and itera- tive solution

    Sören Bartels. “Total variation minimization with finite elements: convergence and itera- tive solution”. In:SIAM Journal on Numerical Analysis50.3 (2012), pp. 1162–1180

    work page 2012

  28. [28]

    Gradient-flow-based semi-implicit finite-element method and its convergence analysis for image reconstruction

    Chong Chen and Guoliang Xu. “Gradient-flow-based semi-implicit finite-element method and its convergence analysis for image reconstruction”. In:Inverse Problems28.3 (2012), p. 035006

    work page 2012

  29. [29]

    Gradient schemes: a generic framework for the discretisation of linear, nonlinear and nonlocal elliptic and parabolic equations

    Jérôme Droniou et al. “Gradient schemes: a generic framework for the discretisation of linear, nonlinear and nonlocal elliptic and parabolic equations”. In:Mathematical Models and Methods in Applied Sciences23.13 (2013), pp. 2395–2432

    work page 2013

  30. [30]

    Discrete total variation flows without regularization

    Sören Bartels, Ricardo H Nochetto, and Abner J Salgado. “Discrete total variation flows without regularization”. In:SIAM Journal on Numerical Analysis52.1 (2014), pp. 363– 385

    work page 2014

  31. [31]

    Linearized FE approximations to a nonlinear gradient flow

    Buyang Li and Weiwei Sun. “Linearized FE approximations to a nonlinear gradient flow”. In:SIAM Journal on Numerical Analysis52.6 (2014), pp. 2623–2646

    work page 2014

  32. [32]

    Error control and adaptivity for a variational model problem defined on functions of bounded variation

    Sören Bartels. “Error control and adaptivity for a variational model problem defined on functions of bounded variation”. In:Mathematics of Computation84.293 (2015), pp. 1217– 1240

    work page 2015

  33. [33]

    Learning nonlinear spectral filters for color image reconstruction

    Michael Moeller et al. “Learning nonlinear spectral filters for color image reconstruction”. In:Proceedings of the IEEE International Conference on Computer Vision. 2015, pp. 289– 297

    work page 2015

  34. [34]

    Unconditional stability of semi- implicit discretizations of singular flows

    Sören Bartels, Lars Diening, and Ricardo H Nochetto. “Unconditional stability of semi- implicit discretizations of singular flows”. In:SIAM Journal on Numerical Analysis56.3 (2018), pp. 1896–1914

    work page 2018

  35. [35]

    Improved estimate for gradient schemes and super- convergence of the TPFA finite volume scheme

    Jérôme Droniou and Neela Nataraj. “Improved estimate for gradient schemes and super- convergence of the TPFA finite volume scheme”. In:IMA Journal of Numerical Analysis 38.3 (2018), pp. 1254–1293

    work page 2018

  36. [36]

    Jérôme Droniou et al.The gradient discretisation method. Vol. 82. Springer, 2018

    work page 2018

  37. [37]

    Runge–Kutta time discretization of nonlinear parabolic equations studied via discrete maximal parabolic regularity

    Peer C. Kunstmann, Buyang Li, and Christian Lubich. “Runge–Kutta time discretization of nonlinear parabolic equations studied via discrete maximal parabolic regularity”. In: Foundations of Computational Mathematics18 (2018), pp. 1109–1130

    work page 2018

  38. [38]

    Crouzeix–Raviart approximation of the total vari- ation on simplicial meshes

    Antonin Chambolle and Thomas Pock. “Crouzeix–Raviart approximation of the total vari- ation on simplicial meshes”. In:Journal of Mathematical Imaging and Vision62 (2020), pp. 872–899

    work page 2020

  39. [39]

    The hybrid high-order method for poly- topal meshes

    Daniele Antonio Di Pietro and Jérôme Droniou. “The hybrid high-order method for poly- topal meshes”. In:Number 19 in Modeling, Simulation and Application84 (2020)

    work page 2020

  40. [40]

    The gradient discretization method for slow and fast diffusion porous media equations

    Jérôme Droniou and Kim-Ngan Le. “The gradient discretization method for slow and fast diffusion porous media equations”. In:SIAM Journal on Numerical Analysis58.3 (2020), pp. 1965–1992

    work page 2020

  41. [41]

    A new numerical scheme for discrete constrained total variation flows and its convergence

    Yoshikazu Giga et al. “A new numerical scheme for discrete constrained total variation flows and its convergence”. In:Numerische Mathematik146.1 (2020), pp. 181–217

    work page 2020

  42. [42]

    A second-order stabilization method for lin- earizing and decoupling nonlinear parabolic systems

    Buyang Li, Yuki Ueda, and Guanyu Zhou. “A second-order stabilization method for lin- earizing and decoupling nonlinear parabolic systems”. In:SIAM Journal on Numerical Analysis58.5 (2020), pp. 2736–2763. 29

    work page 2020

  43. [43]

    Linearization of the finite element method for gradient flows by Newton’s method

    Georgios Akrivis and Buyang Li. “Linearization of the finite element method for gradient flows by Newton’s method”. In:IMA Journal of Numerical Analysis41.2 (2021), pp. 1411– 1440

    work page 2021

  44. [44]

    The convergence of a numerical method for total variation flow

    Qianying Hong, Ming-jun Lai, and Jingyue Wang. “The convergence of a numerical method for total variation flow”. In:Journal of Algorithms & Computational Technol- ogy15 (2021), p. 17483026211011323

    work page 2021

  45. [45]

    Error estimates for total-variation regularized mini- mization problemswith singular dual solutions

    Sören Bartels and Alex Kaltenbach. “Error estimates for total-variation regularized mini- mization problemswith singular dual solutions”. In:Numerische Mathematik152.4 (2022), pp. 881–906

    work page 2022

  46. [46]

    Singular solutions, graded meshes, and adaptivity for total-variation regularized minimization problems

    Sören Bartels, Robert Tovey, and Friedrich Wassmer. “Singular solutions, graded meshes, and adaptivity for total-variation regularized minimization problems”. In:ESAIM: Math- ematical Modelling and Numerical Analysis56.6 (2022), pp. 1871–1888. Appendices A.Auxiliary results LetF ρ :R 2 →Rbe defined viaF ρ(·) =·/| · | ρ with| · | ρ defined in (2.1). Propos...

    work page 2022