Pith. sign in

REVIEW 3 major objections 4 minor 41 references

A Liouville principle for the random conductance model under degenerate conditions

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 The paper proves that in a degenerate random conductance model on $\mathbb{Z}^d$, the space of harmonic functions with sub-$(1+\alpha)$ growth is exactly the $(d+1)$-dimensional space of affine functions.

desk verdict A genuinely new discrete Liouville theorem for degenerate random conductances, but the excess-decay core is explicitly imported from a companion paper and not proved here. read the letter →

arxiv 1908.10691 v2 pith:FJGPBQFQ submitted 2019-08-28 math.AP cs.NAmath.NAmath.PR

Tuan Anh Nguyen This is my paper · ORCID
classification math.APcs.NAmath.NAmath.PR MSC 60K3735B5335J15
keywords randomconductancemodelLiouvilletheoremdegenerateconductanceshomogenizationcorrectorsdiscreteharmonicfunctionsexcessdecayfinite-elementinterpolation
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 a first-order Liouville theorem for the random conductance model on $\mathbb{Z}^d$ when conductances may be arbitrarily large or close to zero. Under stationarity, ergodicity, reflection invariance, and a two-sided moment bound with $1/p+1/q \le 2/d$, the only harmonic functions that grow more slowly than $|x|^{1+\alpha}$ are the affine functions: constants plus the $d$ coordinate functions. The theorem matters because it extends a known continuum result to discrete random media and covers degenerate environments where uniform ellipticity fails, including conductances given by the exponential of a massless Gaussian free field. The proof constructs sublinear correctors and adapts an excess-decay argument to the lattice, with the key boundary regularity estimates assumed from a companion paper.

What carries the argument

The argument runs on three objects: the first-order correctors $\varphi_i$ (harmonic coordinates) and second-order correctors $\sigma_{ijk}$ solving discrete Poisson equations; the excess decay estimate of Corollary 2.3, which controls the homogenization error in a box of radius $R$ by $(r/R)^\alpha$; and a discrete smoothing toolkit built from a finite-element interpolation projection and a dual smoothing operator, which transfers continuum boundary estimates to the lattice. Reflection invariance of the environment makes the homogenized matrix $a^h$ diagonal, so the only affine directions are the $d$ coordinate axes.

What would settle it

Compute, for a reflection-invariant stationary ergodic conductance law satisfying the moment bound, the discrete Dirichlet and Neumann extensions on the box $D_R$ and test whether inequalities (91) and (92) hold with a constant independent of $R$. A sequence of boxes where the normal-to-tangential gradient ratio blows up would falsify the proof's key input; more decisively, exhibiting an $\omega$-harmonic function with $o(R^{1+\alpha})$ growth that is not affine would falsify the theorem itself.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.4: for $\mathbb{P}$-almost every environment $\omega$, the set $S(\omega)$ of $\omega$-harmonic functions $u$ satisfying the growth condition $\lim_{R\to\infty} R^{-(1+\alpha)} (\sum_{|y|_\infty<R} |u(y)|^{2p/(p-1)})^{(p-1)/(2p)} = 0$ is a linear space of dimension exactly $d+1$. Equivalently, every such function is of the form $u(x) = a + b\cdot x$ with $a\in\mathbb{R}$ and $b\in\mathbb{R}^d$. The paper establishes this by constructing first-order and second-order correctors with sublinear growth, proving a deterministic excess decay estimate on boxes, and using reflection invariance to reduce the homogenized matrix to a diagonal matrix. The excess decay proof adapts the continuum argument to the discrete setting, with boundary regularity estimates treated as assumptions supplied by the author's companion paper.

Load-bearing premise

The proof assumes, without proving it here, that on box boundaries the discrete gradients of harmonic extensions in tangential and normal directions are quantitatively comparable, together with a discrete Sobolev inequality; if these boundary regularity estimates fail, the excess-decay argument and hence the theorem are not established.

Editorial extensions

If this is right

  • If Theorem 1.4 is correct, every $\omega$-harmonic function with sub-$(1+\alpha)$ growth is affine, so the medium cannot host non-trivial slowly growing harmonic modes.
  • The result covers conductances with unbounded upper and lower ratios, showing Liouville behavior is stable under degeneracies that satisfy the moment condition.
  • The discrete excess decay and corrector construction provide a route to quantitative homogenization and higher-order Liouville statements in the same degenerate setting.
  • The dimension $d+1$ matches the classical lattice Liouville theorem, so the random medium does not alter the space of harmonic coordinates even when ellipticity constants are absent.

Reading between the lines

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

  • The paper relaxes the strict inequality $1/p+1/q<2/d$ used in the continuum to $1/p+1/q\le 2/d$, which suggests that boundary regularity assumptions, rather than interior estimates, are the true bottleneck; if those assumptions hold under weaker moments, the Liouville dimension may persist beyond the stated range.
  • The method likely extends to non-diagonal homogenized matrices and to random-graph settings once discrete $L^p$ boundary trace theory is developed; the author explicitly notes the non-diagonal case as an open difficulty.
  • A testable consequence is that numerical computation of the dimension of slowly growing harmonic functions on large boxes should return $d+1$ for any reflection-invariant stationary ergodic conductance law satisfying the moment bound; deviations would indicate a missing assumption.
  • The finite-element interpolation and dual smoothing tools developed here could be reused for higher-order Liouville theorems or quantitative homogenization on boxes in degenerate random environments.
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 claims a first-order Liouville theorem for the random conductance model on Z^d with positive but possibly unbounded conductances. Under stationarity, ergodicity, reflection invariance, and moment bounds ||mu_e||_{L^p} + ||mu_e^{-1}||_{L^q} < infinity with 1/p + 1/q <= 2/d, the space of omega-harmonic functions satisfying a sublinear-growth condition of order o(R^{1+alpha}) in an averaged L^{2p/(p-1)} norm is asserted to have dimension d+1. The proof follows the continuum strategy of Bella, Fehrman, and Otto: construct first- and second-order correctors with sublinearity, then combine them with a deterministic excess-decay estimate. The corrector construction in Section 3 is largely self-contained, while the excess decay is based on boundary regularity estimates stated as assumptions in Setting 6.1 and deferred to the companion paper [37].

Significance. If the deferred boundary estimates hold, the theorem is a meaningful extension of known Liouville results: it covers degenerate random conductances that are not bounded away from zero or infinity, requires only stationarity and ergodicity together with reflection invariance, and reaches the endpoint moment condition 1/p + 1/q <= 2/d. The paper also gives a careful discrete treatment of the corrector construction, including a density argument for sublinearity, and a detailed finite-element-style smoothing construction on the discrete boundary. The presentation is honest in flagging the assumptions borrowed from [37], and the central derivation is deterministic and structurally sound. However, the manuscript as submitted is not fully self-contained: the proof of the main theorem depends on quantitative boundary estimates that are not proved here, and the Meyer-type estimate is only sketched.

major comments (3)
  1. [Section 2.2 and Setting 6.1, Eqs. (91), (92), (94)] The proof of Corollary 2.3, and hence of Theorem 1.4, relies on boundary regularity estimates that are assumed but not proved in this manuscript. Corollary 6.10, which controls the boundary term (18), uses Lemmas 6.7 and 6.8, both of which invoke the normal/tangential comparability (91). Corollary 6.13, which controls the annulus term (19), uses Lemma 6.12, which invokes the high-exponent bound (92). Corollary 6.11, used for the interior control of the harmonic extensions, uses the discrete Sobolev inequality (94) through Lemma 6.6. These estimates must hold with constants independent of R and u in the specific norms 2p/(p+1), 2q/(q+1), and max{4p/(p-1),4q/(q-1)}. Section 2.2 explicitly states that they are written in another paper and are considered as assumptions here. If [37] does not prove exactly these inequalities with uniform constants, or if (94) is not valid in the stated form, the excess decay and the main theorem are not established. This is a load-bearing gap that should be resolved by including full proofs or by verifying the exact statements against [37] in an appendix.
  2. [Section 3.2, Lemma 3.5] The Meyer-type estimate for the projection Hf onto L^2_{\nabla^*} is an important step in proving that the second-order corrector belongs to L^{2p/(p+1)} and is sublinear. Its proof is only sketched: it refers to a 'standard argument with good and bad boxes' in [10], uses Green-function derivative bounds from [35], and then says that a duality argument completes the proof. Since this estimate is used to ensure the correctors have the integrability needed for Corollary 2.2, the proof should be written out in full or replaced by a precise statement with a complete derivation. As written, this is a second place where a central ingredient is deferred rather than proved.
  3. [Section 2.2, Corollary 2.3] Corollary 2.3 is stated as a main deterministic ingredient, but its proof is not fully contained in the paper. Section 2.2 sketches the energy estimates and then says that Steps 3, 4, and 5 'can be easily adapted' from the continuum proof of Bella, Fehrman, and Otto. Since the excess decay is the bridge between the sublinear correctors and the Liouville conclusion, the manuscript should provide a complete deterministic argument for Corollary 2.3, including the induction/iteration over scales, rather than relying on an external paper for the final steps. At minimum, the exact discrete analogues of the omitted steps should be stated with their hypotheses and proof.
minor comments (4)
  1. [Corollary 6.11] In the proof of Corollary 6.11(i) the text reads '2c^6 R Lambda(u)' where the statement has '2c^6 Lambda(u)'; the spurious factor R should be removed or the inequality should be restated consistently.
  2. [Introduction, Section 1.1] There is a typo 'haft-space' in the description of the Fischer--Raithel result; it should read 'half-space'.
  3. [Section 2.2, Eq. (13)] The notation in (13) is dense and the free parameter gamma appears both in a favorable and an unfavorable power; a short sentence explaining the final choice of gamma would improve readability.
  4. [References, [37]] Reference [37] is described as 'submitted to Potential Analysis' with an arXiv identifier; since the current paper's main theorem is conditional on results in [37], the author should provide a precise pointer to the theorem or equation numbers in [37] that imply (91), (92), and (94).

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the Liouville theorem is derived from constructed correctors and a deterministic excess decay; the deferred boundary estimates in [37] are an external correctness gap, not a circular reduction.

full rationale

The paper's central claim, Theorem 1.4, is obtained from the existence and sublinearity of the correctors (Corollary 2.2) and a deterministic excess decay (Corollary 2.3), not by assuming the target dimension d+1. The correctors are constructed by Hilbert-space projections and their sublinearity is proved through a density argument and discrete Sobolev inequalities (Lemmas 3.3, 3.4, 3.5, 3.6, 3.8), with no parameter fitted to the desired conclusion. The excess decay is adapted from Bella, Fehrman, and Otto [5] and depends on boundary regularity estimates (91), (92), and Sobolev inequality (94) listed as assumptions in Setting 6.1; the paper explicitly states that these are proved in the author's companion paper [37], and Section 2.2 says: "This contains many technicalities and is written in another paper (see [37]). Here, we consider it as assumptions (see (91) and (92) in Setting 6.1)." This reliance is a genuine external dependency and a correctness risk if the companion estimates fail, but it is not circular: the assumptions are stated, the target Liouville theorem is not among them, and the derivation does not reintroduce the theorem's conclusion as an input. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported to force a choice, and no known result is relabeled as new. Accordingly, the appropriate finding is no significant circularity, with any concern about the deferred boundary estimates belonging to correctness verification rather than circularity analysis.

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

The main theorem does not use fitted parameters. The analysis relies on standard ergodic theorems, Sobolev inequalities, discrete elliptic estimates, and Green function estimates, plus several boundary regularity assumptions that are explicitly deferred to the author's companion paper [37].

assumptions (6)
  • ad hoc to paper Boundary regularity of harmonic extensions: for all u, |nabla(Nu)|_{s,Etau_R} <= c |nabla(Nu)|_{s.Enu_R} and |nabla(Du)|_{s,Enu_R} <= c |nabla(Du)|_{s,Etau_R} for s in {2p/(p+1), 2q/(q+1)}.
    Setting 6.1, inequality (91); stated without proof, deferred to the author's [37]. It is used in Lemmas 6.7-6.8 and Corollary 6.11 to control boundary terms in the excess decay.
  • ad hoc to paper High-exponent boundary regularity: |nabla(Du)|_{max{4p/(p-1),4q/(q-1)}, E_R} <= c |nabla(Du)|_{...,Etau_R} and analogous for N with Enu_R.
    Setting 6.1, inequality (92); also deferred to [37], used in Lemma 6.12 and Corollary 6.13 for the annulus term.
  • ad hoc to paper Sobolev inequality on discrete boundary: inf_a |u-a|_{2r/(r-1), partial D_R} <= c R |nabla u|_{alpha(r), Etau_R} for r in {p,q}.
    Setting 6.1, inequality (94); stated as an assumption, needed for Lemma 6.6 and Corollary 6.11.
  • standard math Existence and stability of smoothing operators S_{R,eps} on discrete surfaces with estimates (84)-(86).
    Corollary 5.8 is proved from the Scott-Zhang construction, but Lemma 5.2 for continuum Lipschitz domains is stated without proof, flagged as technical but standard.
  • standard math Meyer-type L^r estimate for the projection onto L^2_{nabla*}: ||Hf||_{L^r} <= c ||f||_{L^r} for r in (1, infinity).
    Lemma 3.5 is only sketched via good-bad boxes and Green function estimates from [35]; it is used in Corollary 2.2 to get integrability of the correctors.
  • standard math Standard discrete elliptic estimates (mean value inequality, Caccioppoli, maximal inequality) hold for harmonic functions.
    Used throughout Section 2, e.g., equations (20)-(22), with citations to [15], [16], [13], [3]. These are background results, not proved here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Liouville principle for the random conductance model under degenerate conditions." pith.science (2026). https://pith.science/paper/FJGPBQFQ

@misc{pith2026190810691,
  author       = {Pith},
  title        = {Pith review of: A Liouville principle for the random conductance model under degenerate conditions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FJGPBQFQ}},
  note         = {Machine review of arXiv:1908.10691}
}
abstract

We consider a random conductance model on the $d$-dimensional lattice, $d\in[2,\infty)\cap\mathbb{N}$, where the conductances take values in $(0,\infty)$ and are however not assumed to be bounded from above and below. We assume that the law of the random conductances is stationary and ergodic with respect to translations on $\mathbb{Z}^d$ and invariant with respect to reflections on $\mathbb{Z}^d$ and satisfies a similar moment bound as that by Andres, Deuschel, and Slowik (2015), under which a quenched FCLT holds. We prove a first-order Liouville theorem. In the proof we construct the sublinear correctors in the discrete and adapt boundary estimates for harmonic extensions from the work in the continuum done by Bella, Fehrman, and Otto (2018) to the discrete.

Figures

Figures reproduced from arXiv: 1908.10691 by the authors.

Figure 1
Figure 1. Normal and tangential edges in the two-dimensional case Left. The sets Eτ R, Eν R, and ˜∂DR in Setting 2.4 for d = 2, N = 10: Eτ R contains all oriented red edges going in both direction, Eν R contains all exterior blue normal edges, and ˜∂DR contains all red points without four corners ◦. Right. The definition of nR in Setting 2.4: for every y ∈ ˜∂DR there exists a unique x ∈ DR such that (x, y) ∈ Eν R and we set n… view at source ↗
Figure 2
Figure 2. An illustrative construction of a good cover of the surface of a box iii) it holds that |{x ∈ ∂DR |z ∈ Γx}| ≤ K and [PITH_FULL_IMAGE:figures/full_fig_p023_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 40 canonical work pages

  1. [37]

    T. A. Nguyen. An Lp-comparison, p∈ (1,∞), on the finite differences of a discrete harmonic function at the boundary of a discrete box. Submitted to Potential Analysis, arXiv:1905.08151 , 2019

    work page Pith review arXiv 1905

  2. [10]

    Biskup, M

    M. Biskup, M. Salvi, and T. Wolff. A central limit theorem for the effec tive conductance: Linear boundary data and small ellipticity contrasts. Communications in Mathematical Physics , 328:701–731, June 2014

    work page 2014

  3. [35]

    J.-C. Mourrat. First-order expansion of homogenized coefficient s under bernoulli perturbations. Journal de Math´ ematiques Pures et Appliqu´ ees, 103(1):68–101, 2013

    work page 2013

  4. [1]

    Decoupling inequalities and supercritical percolation for the vacant set of random walk loop soup

    C. Alves and A. Sapozhnikov. Decoupling inequalities and supercritic al percolation for the vacant set of random walk loop soup. arXiv:1808.01277, 2018

    work page Pith review arXiv 2018

  5. [2]

    Andres, J.-D

    S. Andres, J.-D. Deuschel, and M. Slowik. Invariance principle for t he random conductance model in a degenerate ergodic environment. The Annals of Probability , 43(4):1866–1891, 2015

    work page 2015

  6. [3]

    Andres, J.-D

    S. Andres, J.-D. Deuschel, and M. Slowik. Harnack inequalities on we ighted graphs and some applications to the random conductance model. Probability Theory Related Fields , 164(3-4):931–977, 2016

    work page 2016

  7. [4]

    Elliptic regularity and quantitative homogenization on percolation clusters

    S. Armstrong and P. Dario. Elliptic regularity and quantitative homo genization on percolation clus- ters. Communications on Pure and Applied Mathematics ( to appear )., 2017. Preprint avaiable on https://arxiv.org/abs/1609.09431

    work page Pith review arXiv 2017

  8. [5]

    Bella, B

    P. Bella, B. Fehrman, and F. Otto. A Liouville theorem for elliptic syst ems with degenerate ergodic coefficients. Annal of Applied Probability , 28(3):1379–1422, 2018

    work page 2018

Show all 41 references
  1. [6]

    Bella and M

    P. Bella and M. Sch¨ affner. Quenched invariance principle for rando m walks among random degenerate conductances. arXiv:1902.05793, 2019

    1902 arXiv

  2. [7]

    Benjamini, H

    I. Benjamini, H. Duminil-Copin, G. Kozma, and A. Yadin. Disorder, en tropy and harmonic functions. The Annals of Probability , 43(5):2332–2373, 2015

    2015

  3. [8]

    Berger and M

    N. Berger and M. Biskup. Quenched invariance principle for simple ra ndom walk on percolation clusters. Probability Theory Related Fields , 137(1-2):83–120, 2007

    2007

  4. [9]

    M. Biskup. Recent progress on the random conductance model. Probability Surveys, 8:294–373, 2011

    2011

  5. [11]

    Brydges and T

    D. Brydges and T. Spencer. Fluctuation estimates for sub-quad ratic gradient field actions. Journal of Mathematical Physics, 53(9):095216, 2012

    2012

  6. [12]

    Buhovsky, A

    L. Buhovsky, A. Logunov, E. Malinnikova, and M. Sodin. A discrete harmonic function bounded on a large portion of Z2 is constant. arXiv:1712.07902, 2017

    2017 arXiv

  7. [13]

    Coulhon and A

    T. Coulhon and A. Grigoryan. Random walks on graphs with regular v olume growth. Geometric and Functional Analysis, 8:656—-701, 1998. LIOUVILLE PRINCIPLE FOR DEGENERATE RCM 33

    1998

  8. [14]

    De Masi, P

    A. De Masi, P. A. Ferrari, S. Goldstein, and W. D. Wick. An invariance principle for reversible Markov processes. Applications to random motions in random environments . Journal of Statistical Physics , 55(3- 4):787–855, 1989

    1989

  9. [15]

    Delmotte

    T. Delmotte. Harnack inequalities on graphs. S´ eminaire de th´ eorie spectrale et g´ eom´ etrie, 16:217–228, 1997-1998

    1997

  10. [16]

    Delmotte and J.-D

    T. Delmotte and J.-D. Deuschel. On estimating the derivatives of sy mmetric diffusions in stationary random environment, with applications to ∇φ interface model. Probability Theory and Related Fields , 133(3):358– 390, Nov. 2005

    2005

  11. [17]

    Deuschel, G

    J.-D. Deuschel, G. Giacomin, and D. Ioffe. Large deviations and con centration properties for ∇φ interface models. Probability Theory and Related Fields , 117(1):49–111, 2000

    2000

  12. [18]

    Deuschel, T

    J.-D. Deuschel, T. A. Nguyen, and M. Slowik. Quenched invariance p rinciples for the random conductance model on a random graph with degenerate ergodic weights. Probability Theory and Related Fields , 170(1- 2):363—-386, Feb. 2017

    2017

  13. [19]

    R. J. Duffin. Discrete potential theory. Duke Mathematical Journal , 2(20):233–251, 1953

    1953

  14. [20]

    Fischer and F

    J. Fischer and F. Otto. A higher-order large-scale regularity the ory for random elliptic operators. Commu- nications in Partial Differential Equations , 41, 2016

    2016

  15. [21]

    Fischer and C

    J. Fischer and C. Raithel. Liouville principles and a large-scale regular ity theory for random elliptic oper- ators on the half-space. SIAM Journal on Mathematical Analysis , 49(1):82–114, 2016

    2016

  16. [22]

    Gloria, S

    A. Gloria, S. Neukamm, and F. Otto. A regularity theory for rando m elliptic operators. Preprint, 2014. Avaiable on https://arxiv.org/abs/1409.2678

    2014 arXiv

  17. [23]

    Gloria and F

    A. Gloria and F. Otto. An optimal variance estimate in stochastic ho mogenization of discrete elliptic equations. The Annals of Probability , 39(3):779–856, 05 2011

    2011

  18. [24]

    Gloria and F

    A. Gloria and F. Otto. An optimal error estimate in stochastic homo genization of discrete elliptic equations. The Annals of Applied Probability , 22(1):1–28, 2012

    2012

  19. [25]

    G¨ urlebeck and A

    K. G¨ urlebeck and A. Hommel. On finite difference dirac operators a nd their fundamental solutions. Ad- vances in Applied Clifford Algebras , 2001

    2001

  20. [26]

    G¨ urlebeck and A

    K. G¨ urlebeck and A. Hommel. On finite difference potentials and the ir applications in a discrete function theory. Mathematical Methods in Applied Sciences , 25:1563–1576, 2002

    2002

  21. [27]

    H. A. Heilbronn. On discrete harmonic functions. Mathematical Proceedings of the Cambridge Philosophical Society, 45(2):194–206, April 1949

    1949

  22. [28]

    A. Hommel. Fundamentall¨ osungen partieller Differenzenoperatoren und die L¨ osung diskreter Randwert- probleme mit Hilfe von Differenzenpotentialen . PhD thesis, Bauhaus-Universit¨ at Weimar, 1998

    1998

  23. [29]

    K¨ unnemann

    R. K¨ unnemann. The diffusion limit for reversible jump processes on zd with ergodic random bond conduc- tivities. Communications in Mathematical Physics , 90(1):27–68, 1983

    1983

  24. [30]

    H. Lewy, K. Friedrichs, and R. Courant. ¨Uber die partiellen Differenzengleichungen der mathematischen Physik. Mathematische Annalen , 100:32–74, 1928

    1928

  25. [31]

    Lippner and D

    G. Lippner and D. Mangoubi. Harmonic functions on the lattice: Abs olute monotonicity and propagation of smallness. Duke Mathematical Journal , 164(13):2577–2595, 2015

    2015

  26. [32]

    Lippner and D

    G. Lippner and D. Mangoubi. On the sharpness of a three circles th eorem for discrete harmonic functions. arXiv:1512.03732, 2015

    2015 arXiv

  27. [33]

    Logunov and E

    A. Logunov and E. Malinnikova. Ratios of harmonic functions with th e same zero set. arXiv:1506.08041, 2015

    2015 arXiv

  28. [34]

    Mitrea and M

    I. Mitrea and M. Mitrea. Multi-Layer Potentials and Boundary Problems for Higher-O rder Elliptic Systems in Lipschitz Domains . Springer, 2013

    2013

  29. [36]

    T. A. Nguyen. The random conductance model under degenerate conditions . PhD thesis, Technische Uni- versit¨ at Berlin, 2017

    2017

  30. [38]

    Rodriguez

    P.-F. Rodriguez. Decoupling inequalities for the Ginzburg-Landau ∇ϕ models. arXiv:1612.02385, 2016

    2016 arXiv

  31. [39]

    Sapozhnikov

    A. Sapozhnikov. Random walks on infinite percolation clusters in mod els with long-range correlations. The Annals of Probability , 45(3):1842–1898, 2016

    2016

  32. [40]

    L. R. Scott and S. Zhang. Finite element interpolation of nonsmoot h functions satisfying boundary condi- tions. Mathematics of Computation , 54:483–493, 1990

    1990

  33. [41]

    F. Stummel. Elliptische Differenzenoperatoren unter Dirichletrand bedingungen. Mathematische Zeitschrift, 97:169–211, 1967

    1967

Pith tools

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