pith. sign in

arxiv: 1906.08700 · v1 · pith:UNWR6JULnew · submitted 2019-06-20 · 🧮 math.AP · cs.NA· math.NA

On quasi-reversibility solutions to the Cauchy problem for the Laplace equation: regularity and error estimates

Laurent Bourgeois , Lucas Chesnel This is my paper

Pith reviewed 2026-05-25 19:11 UTC · model grok-4.3

classification 🧮 math.AP cs.NAmath.NA
keywords Cauchy problemLaplace equationquasi-reversibilityregularity estimatesKondratiev theorypolygonal domainsill-posed problemsfinite elements
0
0 comments X

The pith

Quasi-reversibility approximations to the ill-posed Laplace Cauchy problem admit regularity estimates uniform in the regularization parameter.

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

The paper shows that a family of regularized well-posed problems, introduced to approximate solutions of the Cauchy problem for the Laplace equation, possess regularity properties that remain controlled as the small parameter ε tends to zero. These uniform estimates are required to justify convergence of finite-element discretizations of the regularized problems. The results are proved first in smooth domains and then extended to two-dimensional polygonal domains. In the polygonal case the Kondratiev theory is applied because the usual Grisvard technique fails on the regularized operators, and every constant and singularity exponent is tracked explicitly with respect to ε. The central feature is that all estimates remain valid even though the original Cauchy problem is ill-posed in every functional setting.

Core claim

We obtain regularity results uniform in ε for the solutions to the regularized problems both in smooth domains and in 2D polygonal geometries. In polygonal domains the Kondratiev theory is applied to control the dependence on ε in the estimates, even though the limit problem is ill-posed in any framework.

What carries the argument

The one-parameter family of regularized elliptic problems whose solutions are analyzed by the Kondratiev theory so that singularity exponents and constants stay bounded independently of ε.

If this is right

  • Finite-element methods applied to the regularized problems converge with rates independent of ε.
  • Error bounds between the regularized solution and any exact solution of the original Cauchy problem can be obtained once the mesh size tends to zero.
  • The same Kondratiev-based estimates remain valid when the data are only compatible in a weak sense.
  • The approach extends directly to other linear elliptic operators whose principal part is the Laplacian.

Where Pith is reading between the lines

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

  • The uniform control on singularity exponents suggests that the same regularization can be used for the Cauchy problem on three-dimensional polyhedra if the corresponding Kondratiev constants remain bounded.
  • One could derive an explicit rate of convergence in ε by combining the present regularity statements with a separate stability estimate for the regularized operator.
  • The method may supply a template for obtaining uniform regularity in other ill-posed boundary-value problems that admit a similar quasi-reversibility regularization.

Load-bearing premise

The Kondratiev theory applies to the regularized problems with constants and singularity exponents that stay controlled uniformly as ε approaches zero.

What would settle it

A calculation in a concrete polygonal domain showing that the Sobolev norms of the regularized solutions become unbounded as ε tends to zero would falsify the uniform regularity claim.

Figures

Figures reproduced from arXiv: 1906.08700 by Laurent Bourgeois, Lucas Chesnel.

Figure 1
Figure 1. Figure 1: Examples of domains Ω. The thick blue lines represe [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: An example of polygonal domain. S1, S2, S3 represent the three types of vertices that we will study in §3.2, §5, §3.3 respectively. assume that ∂Ω is the union of segments Γj , j = 1, . . . , N, where N is an integer. Let us denote Sj the vertex such that Sj = Γj ∩ Γj+1, ωj the angle between Γj and Γj+1 from the interior of Ω, τj the unit tangent oriented in the counter-clockwise sense and νj the outward n… view at source ↗
Figure 3
Figure 3. Figure 3: Position of the λ ± n in the complex plane. Lemma 5.1. The eigenvalues of the symbol Lε are λ ± n = 1 ω  π 2 + nπ ± iln γε  , n ∈ Z, with γε = r 1 + 1 ε + r 1 ε (see [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
read the original abstract

We are interested in the classical ill-posed Cauchy problem for the Laplace equation. One method to approximate the solution associated with compatible data consists in considering a family of regularized well-posed problems depending on a small parameter $\varepsilon>0$. In this context, in order to prove convergence of finite elements methods, it is necessary to get regularity results of the solutions to these regularized problems which hold uniformly in $\varepsilon$. In the present work, we obtain these results in smooth domains and in 2D polygonal geometries. In presence of corners, due the particular structure of the regularized problems, classical techniques \`a la Grisvard do not work and instead, we apply the Kondratiev approach. We describe the procedure in detail to keep track of the dependence in $\varepsilon$ in all the estimates. The main originality of this study lies in the fact that the limit problem is ill-posed in any framework.

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 develops regularity theory for a family of quasi-reversibility regularizations (parameter ε>0) of the ill-posed Cauchy problem for the Laplace equation. It proves Sobolev and weighted Sobolev estimates that are uniform in ε, first in smooth domains and then in 2D polygonal domains; in the latter setting the authors replace Grisvard-type arguments by a Kondratiev/Mellin-symbol analysis whose constants and singularity exponents are tracked explicitly with respect to ε. The goal is to furnish the a-priori bounds needed for convergence analysis of finite-element discretizations of the regularized problems.

Significance. If the claimed uniformity of the Kondratiev constants and exponents holds, the work supplies a concrete analytic foundation for the numerical approximation of a classically ill-posed elliptic problem in domains with corners—an area where standard elliptic theory breaks down for the unregularized limit. The explicit tracking of ε-dependence is a positive feature that directly supports passage to the limit in discretization error estimates.

major comments (2)
  1. [§4] §4 (Kondratiev analysis for the regularized operator): the indicial equation associated with the ε-dependent regularized Laplacian is not shown to have roots whose real parts remain bounded away from the integers uniformly down to ε=0. Because the Mellin symbol itself depends on ε, any ε-dependent drift of a root toward an integer would make the weighted-norm constants blow up, undermining the uniformity asserted for the polygonal-domain estimates and the subsequent FEM convergence argument.
  2. [Theorem 5.1] Theorem 5.1 (or the main uniform regularity statement): the proof sketch invokes standard Kondratiev theory applied to each fixed-ε problem, yet does not supply an explicit lower bound on the distance of the indicial roots to the critical line that is independent of ε. Without such a bound the passage from the regularized solutions to the claimed ε-uniform estimates is not justified.
minor comments (2)
  1. [Abstract / §1] The abstract and introduction repeatedly state that 'the procedure is described in detail to keep track of the dependence in ε'; a short table or remark collecting the ε-independent quantities (e.g., the minimal distance of indicial roots) would make this tracking easier to verify.
  2. [§2] Notation for the regularized boundary conditions (the precise form of the ε-terms) should be fixed once at the beginning of §2 rather than re-introduced in each subsequent section.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. The points raised concern the justification of ε-uniformity in the Kondratiev constants for the polygonal case; we address them directly below and will strengthen the manuscript accordingly.

read point-by-point responses

  1. Referee: [§4] §4 (Kondratiev analysis for the regularized operator): the indicial equation associated with the ε-dependent regularized Laplacian is not shown to have roots whose real parts remain bounded away from the integers uniformly down to ε=0. Because the Mellin symbol itself depends on ε, any ε-dependent drift of a root toward an integer would make the weighted-norm constants blow up, undermining the uniformity asserted for the polygonal-domain estimates and the subsequent FEM convergence argument.

    Authors: We agree that an explicit verification is needed. The regularized operator yields a fourth-order Mellin symbol whose coefficients depend continuously on ε. The indicial roots therefore depend continuously on ε. In the limit ε→0 the symbol reduces to that of the Laplacian, whose roots lie at non-integer locations for the angles under consideration. By continuity, for all sufficiently small ε the distance to the nearest integer remains bounded below by a positive constant independent of ε; for ε bounded away from zero the bound is immediate by compactness. We will insert a short lemma in the revised §4 that records this separation explicitly and recomputes the weighted-norm constants with the ε-independent lower bound. revision: yes

  2. Referee: [Theorem 5.1] Theorem 5.1 (or the main uniform regularity statement): the proof sketch invokes standard Kondratiev theory applied to each fixed-ε problem, yet does not supply an explicit lower bound on the distance of the indicial roots to the critical line that is independent of ε. Without such a bound the passage from the regularized solutions to the claimed ε-uniform estimates is not justified.

    Authors: The observation is correct: the current sketch of Theorem 5.1 applies the standard Kondratiev theory for each fixed ε but does not yet display the ε-independent separation constant. The lemma we will add to §4 supplies precisely this lower bound. Once inserted, the application of the theory becomes uniform in ε and justifies the claimed estimates. We will also update the statement of Theorem 5.1 to reference the new lemma. revision: yes

Circularity Check

0 steps flagged

No circularity; standard Kondratiev theory applied with explicit ε-tracking to a new ill-posed setting.

full rationale

The derivation applies classical Kondratiev weighted Sobolev estimates to a family of regularized well-posed problems, with the paper stating that it tracks ε-dependence explicitly in all constants and exponents to obtain uniformity as ε→0. No equations reduce by construction to fitted inputs, self-definitions, or prior self-citations; the limit problem is acknowledged as ill-posed, and the analysis rests on standard elliptic theory without renaming known results or smuggling ansatzes. The central uniformity claim is presented as a direct consequence of the detailed tracking procedure rather than an input.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper relies on background elliptic regularity theory and the well-posedness of each regularized problem; no free parameters or new entities are introduced.

axioms (2)
  • standard math Standard Sobolev-space elliptic regularity and trace theorems hold in the domains under consideration.
    Invoked to obtain a priori estimates for the regularized problems.
  • domain assumption The quasi-reversibility regularization yields a well-posed problem for every fixed ε > 0.
    Stated as the starting point of the approximation strategy.

pith-pipeline@v0.9.0 · 5694 in / 1386 out tokens · 32483 ms · 2026-05-25T19:11:14.165816+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

40 extracted references · 40 canonical work pages · 1 internal anchor

  1. [1]

    Alessandrini, L

    G. Alessandrini, L. Rondi, E. Rosset, and S. Vessella. Th e stability for the Cauchy problem for elliptic equations. Inverse Problems, 25(12):123004, 47, 2009

    work page 2009

  2. [2]

    Azaïez, F.B

    M. Azaïez, F.B. Belgacem, and H. El Fekih. On Cauchy’s pro blem: II. Completion, regular- ization and approximation. Inverse Problems, 22(4):1307, 2006

    work page 2006

  3. [3]

    Belgacem

    F.B. Belgacem. Why is the Cauchy problem severely ill-po sed? Inverse Problems , 23(2):823– 836, 2007

    work page 2007

  4. [4]

    Bourgeois

    L. Bourgeois. A mixed formulation of quasi-reversibili ty to solve the Cauchy problem for Laplace’s equation. Inverse Problems, 21(3):1087–1104, 2005

    work page 2005

  5. [5]

    Bourgeois

    L. Bourgeois. About stability and regularization of ill -posed elliptic Cauchy problems: the case of C 1, 1 domains. M2AN Math. Model. Numer. Anal. , 44(4):715–735, 2010

    work page 2010

  6. [6]

    Bourgeois and J

    L. Bourgeois and J. Dardé. About stability and regulariz ation of ill-posed elliptic Cauchy problems: the case of Lipschitz domains. Appl. Anal., 89(11):1745–1768, 2010

    work page 2010

  7. [7]

    Bourgeois and A

    L. Bourgeois and A. Recoquillay. A mixed formulation of t he Tikhonov regularization and its application to inverse PDE problems. ESAIM Math. Model. Numer. Anal. , 52(1):123–145, 2018

    work page 2018

  8. [8]

    H. Brezis. Analyse fonctionnelle : théorie et applications . Editions Dunod, 1999

    work page 1999

  9. [9]

    E. Burman. Stabilized finite element methods for nonsymm etric, noncoercive, and ill-posed problems. Part I: Elliptic equations. SIAM J Sci Comput. , 35(6):A2752–A2780, 2013

    work page 2013

  10. [10]

    E. Burman. Stabilized finite element methods for nonsym metric, noncoercive, and ill-posed problems. Part II: Hyperbolic equations. SIAM J Sci Comput. , 36(4):A1911–A1936, 2014

    work page 2014

  11. [11]

    E. Burman. A stabilized nonconforming finite element me thod for the elliptic Cauchy problem. Math. Comput. , 86(303):75–96, 2017

    work page 2017

  12. [12]

    Burman, P

    E. Burman, P. Hansbo, and M.G. Larson. Solving ill-pose d control problems by stabilized finite element methods: an alternative to Tikhonov regularizatio n. Inverse Problems , 34(3):035004, 2018. 31

    work page 2018

  13. [13]

    Burman, M.G

    E. Burman, M.G. Larson, and L. Oksanen. Primal-dual mix ed finite element methods for the elliptic Cauchy problem. SIAM J. Numer. Anal. , 56(6):3480–3509, 2018

    work page 2018

  14. [14]

    Chesnel and P

    L. Chesnel and P. Ciarlet Jr. T-coercivity and continuous Galerkin methods: application to transmission problems with sign changing coefficients. Numer. Math. , 124(1):1–29, 2013

    work page 2013

  15. [15]

    Chesnel, X

    L. Chesnel, X. Claeys, and S.A. Nazarov. A curious insta bility phenomenon for a rounded corner in presence of a negative material. Asymp. Anal., 88(1):43–74, 2014

    work page 2014

  16. [16]

    Chesnel, X

    L. Chesnel, X. Claeys, and S.A. Nazarov. Oscillating be haviour of the spectrum for a plasmonic problem in a domain with a rounded corner. Math. Mod. Num. Anal. , 52(4):1285–1313, 2018

    work page 2018

  17. [17]

    P.G. Ciarlet. The finite element method for elliptic problems . North-Holland Publishing Co., Amsterdam-New York-Oxford, 1978. Studies in Mathematics a nd its Applications, Vol. 4

    work page 1978

  18. [18]

    Costabel and M

    M. Costabel and M. Dauge. A singularly perturbed mixed b oundary value problem. Comm. Partial Differential Equations , 21(11-12):1919–1949, 1996

    work page 1919

  19. [19]

    J. Dardé. Iterated quasi-reversibility method applie d to elliptic and parabolic data completion problems. Inverse Probl. Imaging , 10(2):379–407, 2016

    work page 2016

  20. [20]

    Dardé, A

    J. Dardé, A. Hannukainen, and N. Hyvönen. An Hdiv-based mixed quasi-reversibility method for solving elliptic Cauchy problems. SIAM J. Numer. Anal. , 51(4):2123–2148, 2013

    work page 2013

  21. [21]

    Grisvard

    P. Grisvard. Elliptic Problems in Nonsmooth Domains . Pitman, London, 1985

    work page 1985

  22. [22]

    Grisvard

    P. Grisvard. Singularities in boundary value problems , volume 22 of Recherches en Mathéma- tiques Appliquées [Research in Applied Mathematics] . Masson, Paris; Springer-Verlag, Berlin, 1992

    work page 1992

  23. [23]

    Hadamard

    J. Hadamard. Sur les problèmes aux dérivées partielles et leur signification physique. Princeton university bulletin , pages 49–52, 1902

    work page 1902

  24. [24]

    A. M. Il’in. Matching of asymptotic expansions of solutions of boundary value problems, volume 102 of Translations of Mathematical Monographs . AMS, Providence, RI, 1992

    work page 1992

  25. [25]

    A. Kirsch. The Robin problem for the Helmholtz equation as a singular perturbation problem. Numer. Func. Anal. Opt. , 8(1-2):1–20, 1985

    work page 1985

  26. [26]

    Klibanov and F

    M.V. Klibanov and F. Santosa. A computational quasi-re versibility method for Cauchy prob- lems for Laplace’s equation. SIAM J. Appl. Math. , 51(6):1653–1675, 1991

    work page 1991

  27. [27]

    Kondratiev

    V.A. Kondratiev. Boundary-value problems for ellipti c equations in domains with conical or angular points. Trans. Moscow Math. Soc. , 16:227–313, 1967

    work page 1967

  28. [28]

    V. A. Kozlov, V. G. Mazya, and J. Rossmann. Elliptic boundary value problems in domains with point singularities , volume 52 of Mathematical Surveys and Monographs . American Math- ematical Society, Providence, RI, 1997

    work page 1997

  29. [29]

    Kozlov, V.G

    V.A. Kozlov, V.G. Maz’ya, and J. Rossmann. Elliptic Boundary Value Problems in Domains with Point Singularities , volume 52 of Mathematical Surveys and Monographs . AMS, Provi- dence, 1997

    work page 1997

  30. [30]

    Kozlov, V.G

    V.A. Kozlov, V.G. Maz’ya, and J. Rossmann. Spectral problems associated with corner singu- larities of solutions to elliptic equations , volume 85 of Mathematical Surveys and Monographs . AMS, Providence, 2001

    work page 2001

  31. [31]

    Lattès and J.-L

    R. Lattès and J.-L. Lions. Méthode de quasi-réversibilité et applications . Travaux et Recherches Mathématiques, No. 15. Dunod, Paris, 1967

    work page 1967

  32. [32]

    Maz’ya, S.A

    V.G. Maz’ya, S.A. Nazarov, and B.A. Plamenevski ˘ ı. Asymptotic theory of elliptic boundary value problems in singularly perturbed domains, Vol. 1 . Birkhäuser, Basel, 2000. Translated from the original German 1991 edition

    work page 2000

  33. [33]

    Maz’ya and B.A

    V.G. Maz’ya and B.A. Plamenevski ˘ ı. On the coefficients i n the asymptotics of solutions of elliptic boundary value problems with conical points. Math. Nachr. , 76:29–60, 1977. Engl. transl. Amer. Math. Soc. Transl. 123:57–89, 1984

    work page 1977

  34. [34]

    Nazarov and B.A

    S.A. Nazarov and B.A. Plamenevski ˘ ı. Elliptic problems in domains with piecewise smooth boundaries, volume 13 of Expositions in Mathematics . De Gruyter, Berlin, Germany, 1994

    work page 1994

  35. [35]

    Nazarov and N

    S.A. Nazarov and N. Popoff. Self-adjoint and skew-symme tric extensions of the Laplacian with singular Robin boundary condition. C. R. Acad. Sci. Paris, Ser. I , 356(9):927–932, 2018

    work page 2018

  36. [36]

    Plummeting and blinking eigenvalues of the Robin Laplacian in a cuspidal domain

    S.A. Nazarov, N. Popoff, and J. Taskinen. Plummeting and blinking eigenvalues of the Robin Laplacian in a cuspidal domain. arXiv preprint arXiv:1809.10963 , 2018. 32

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  37. [37]

    S. Nicaise. Regularity of the solutions of elliptic sys tems in polyhedral domains. Bull. Belg. Math. Soc. Simon Stevin , 4(3):411–429, 1997

    work page 1997

  38. [38]

    L.E. Payne. Bounds in the Cauchy problem for the Laplace equation. Arch. Ration. Mech. Anal., 5(1):35–45, 1960

    work page 1960

  39. [39]

    L.E. Payne. On a priori bounds in the Cauchy problem for e lliptic equations. SIAM J. Math. Anal., 1(1):82–89, 1970

    work page 1970

  40. [40]

    Remarques sur l’observabilité pour l’ équation de Laplace

    Kim-Dang Phung. Remarques sur l’observabilité pour l’ équation de Laplace. ESAIM Control Optim. Calc. Var. , 9:621–635, 2003. 33

    work page 2003