Geometric aspects of the Harnack Inequality for a nonlocal heat equation

Mateusz Dembny; Miko{\l}aj Sier\.z\k{e}ga

REVIEW 3 major objections 4 minor 8 references

Geometric aspects of the Harnack Inequality for a nonlocal heat equation

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper proves that the kernel-ratio expression behind a sharp fractional-heat Harnack bound is the chord ratio of a circle, in every dimension.

desk verdict Correct algebraic identity overclaimed as an n-dimensional Harnack generalization; the solution bound is never proved. read the letter →

arxiv 2506.08187 v1 pith:NBJQ3ZA4 submitted 2025-06-09 math.AP

Mateusz Dembny , Miko{l}aj Sierżęga This is my paper

classification math.AP MSC 35R1135K08

keywords fractionalheatequationHarnackinequalityPoissonkernelnonlocaldiffusionWidderrepresentationtheoremhalf-LaplaciancirculargeometryLi-Yau

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 establishes a geometric identity for the fractional heat equation's Poisson kernel in n spatial dimensions: for any two spacetime points A and B, the chord ratio on the circle through A and B equals a power of a product of kernel ratios evaluated at the circle's intersections with t=0. The identity is meant to explain and generalize the geometric reading of the sharp one-dimensional Harnack bound obtained in the authors' earlier work. A sympathetic reader would take the paper's contribution to be the extraction of this clean geometric formula from the earlier technical estimate. The paper does not itself prove an n-dimensional Harnack inequality for solutions; it proves the kernel/chord identity and presents it as the geometric core that a higher-dimensional Harnack argument would use.

What carries the argument

The central object is the circle through $A$ and $B$ whose centre lies on $t=0$ and which lies in the vertical plane spanned by $A$, $B$ and their spatial projections. Its intersections with $t=0$ give the two points $x_*$ and $x^*$, and the chord distances from $A$ and $B$ to those points convert, via the Pythagorean theorem, into ratios of the Poisson kernel $K(x,t;y)=c_n t/(t^2+|x-y|^2)^{(n+1)/2}$. The exponent $(n+1)/2$ in the kernel is exactly what makes the chord ratio equal to the $1/(n+1)$ power of the kernel-ratio product.

What would settle it

Exhibit a positive classical solution $u$ of the n-dimensional fractional heat equation, for example in $n=2$ with explicit initial data, such that for some pair $A,B$ the solution ratio $u(B)/u(A)$ is not controlled by the kernel ratios formed at the circle's intersection points $x_*$ and $x^*$; alternatively, show that some positive classical solution cannot be written as a convolution of the Poisson kernel with its initial data. Either would sever the link between identity (3.1) and a Harnack inequality while leaving the algebraic identity itself untouched.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is Theorem 3.1: for points $A$ and $B$ in $\mathbb{R}^n\times\mathbb{R}_+$, with projections $A'$ and $B'$ on the plane $t=0$, there is a circle lying in the vertical plane through $A$ and $B$, with centre on $t=0$, and its intersection points $x_*$ and $x^*$ with $t=0$ satisfy $D_{20}D_{13}/(D_{10}D_{23}) = (K(A,x_*)K(B,x^*)/(K(B,x_*)K(A,x^*)))^{1/(n+1)}$. The chord lengths $D_{20}, D_{13}, D_{10}, D_{23}$ are Euclidean distances from the two spacetime points to the two intersection points, and $K$ is the fractional heat kernel (Poisson kernel). This identifies the logarithm of the kernel-ratio expression with hyperbolic arc length on the circle, matching the structure of the sharp Harnack ratio from the one-dimensional case.

Load-bearing premise

The transfer from the chord/kernel identity to a Harnack bound for actual solutions assumes that in n dimensions every positive classical solution of the fractional heat equation is a convolution of the Poisson kernel with its initial data; the paper cites this representation only in one dimension and does not state or prove the n-dimensional version.

Editorial extensions

If this is right

For $n=2$, formula (3.1) says the chord ratio is the cube root of $K(A,x_*)K(B,x^*)/(K(B,x_*)K(A,x^*))$, giving an explicit geometric expression for the kernel ratio that would appear in a two-dimensional Harnack bound.
The identity replaces the choice of a connecting curve in classical parabolic Harnack arguments with the arc of a circle through the two points, identified geometrically by the construction.
If an n-dimensional Widder representation theorem becomes available, the same two points $x_*$ and $x^*$ define the natural two-sided bound for $u(B)/u(A)$, generalizing equation (1.4) of the previous paper.
Setting $n=1$ recovers the semicircle picture and the square-root exponent of the original one-dimensional result.

Reading between the lines

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

The abstract's claim of a sharp double-sided Harnack bound in higher dimensions goes beyond the theorem actually proved: Theorem 3.1 concerns only the Poisson kernel, not the solution $u$, so the n-dimensional Harnack inequality itself is still an open step.
A concrete numerical test in two dimensions could decide whether the geometric identity is the right seed for that inequality: for explicit positive solutions, check whether the solution ratio $u(B)/u(A)$ is bracketed by the two kernel ratios formed at $x_*$ and $x^*$.
The same chord/kernel algebra would work for any kernel with the same self-similar homogeneity, so a similar circular picture may hold for other nonlocal operators with matching scaling.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Colleague,

The short version: this paper proves a correct but essentially algebraic identity (Thm 3.1) and presents it as an n-dimensional Harnack generalization that the paper never actually proves. The advertised double-sided bound for positive solutions in n dimensions is not stated as a theorem and no argument is given to transfer the kernel/chord identity into an inequality on solution ratios.

What is actually new: the observation in Section 2 that in 1D the sharp Harnack ratio from [2] can be read as a chord ratio on a circle, and that the input to the hyperbolic distance is a product of Poisson kernel values. That is a genuine geometric insight, and the 2D example (Sections 2.2-2.4) shows how the same chord computation goes through with the 2D kernel. The algebra is correct as far as it goes.

The soft spot is the gap between the identity and a Harnack bound. In 1D the Harnack inequality in [2] rests on Widder's representation and on an extremal argument that pins the solution ratio between the two kernel ratios. For n dimensions the paper supplies neither: the n-dimensional Widder representation is never stated (not even cited), and Section 4 explicitly frames the higher-dimensional steps as a 'conjecture'/'postulate' and says the authors 'bypass extensive computations.' So Theorem 3.1 is a kernel fact, not a solution fact. The chord ratio equals a product of Poisson kernel ratios by direct substitution; no estimate involving u enters.

I do not think the paper is incoherent. It is honest in the discussion about the missing piece. But the abstract overstates: 'establish a connection between a sharp double-sided Harnack bound ... and circular geometry' is only true if you read it as referring to kernel ratios, not solution bounds. As a Harnack generalization, this is not there. The right venue for this material would be a remark or appendix to their previous paper, not a standalone theorem.

Recommendation: desk reject. The identity could be useful to someone working on fractional heat kernel estimates, but the advertised result is absent. No serious referee time needed.

Best,

Referee Report

3 major / 4 minor

Summary. The manuscript claims to connect a sharp double-sided Harnack bound for positive solutions of the fractional heat equation to circular geometry in higher dimensions. After reviewing the one-dimensional result from the authors' previous work, the paper constructs, for two spacetime points A and B in R^n × R_+, a circle in the vertical plane through A, B, and their projections onto t = 0, and proves (Theorem 3.1) that the chord ratio D20 D13 / (D10 D23) equals the (1/(n+1))-power of a product of Poisson kernel ratios at the two intersection points of the circle with the t = 0 hyperplane. Section 4 discusses the relation to the earlier one-dimensional Harnack bound and states that the higher-dimensional argument relies on a postulate. No theorem in the paper asserts an inequality for u(B)/u(A) in n dimensions.

Significance. If the algebraic identity (3.1) were all that is claimed, the note would be a modest but correct observation. However, the advertised contribution—a higher-dimensional generalization of the sharp Harnack bound—is not established: Theorem 3.1 contains no solution u and no inequality. The transfer from kernel ratios to solution ratios would require an n-dimensional Widder representation and an extremal estimate on kernel ratios, neither of which appears. The paper's own Section 4 describes the higher-dimensional steps as a postulate. The one-dimensional predecessor (Theorem 1.3) is a genuine theorem; the present manuscript does not generalize it. The correct algebraic identity does not compensate for the missing analytic estimate, so the significance of the paper as it stands is low.

major comments (3)

[Abstract and Theorem 3.1] The abstract claims "a connection between a sharp double-sided Harnack bound for positive solutions of a fractional heat equation and the circular geometry in higher dimensions." The central n-dimensional result, Theorem 3.1, is an identity for the Poisson kernel and chord lengths: it contains no positive solution u and no inequality, so it does not establish any Harnack bound in n dimensions. The only solution-level Harnack statement in the paper is the one-dimensional Theorem 1.3 quoted from the authors' previous work.
[Section 4] Section 4 explicitly says that the existence of the circle was "postulated" and that the geometric insight allowed the authors to "bypass the extensive computations." This is an admission that the higher-dimensional extension is a conjecture, not a theorem. To upgrade the kernel identity (3.1) to a Harnack inequality for u(B)/u(A), one needs (i) an n-dimensional Widder representation theorem expressing positive classical solutions as convolutions with the Poisson kernel, and (ii) a uniform estimate showing that for every source point the kernel-ratio product lies between the two endpoint values. Neither is stated or proved; the citation of [1] is for one dimension and is not applied.
[Section 3, Eq. (3.1)] Equation (3.1) is a definitional identity: the chord lengths D_ab are exactly the Euclidean distances that appear in the Poisson kernel denominators, so both sides are the same product of distances raised to an exponent. The identity is forced by the notation and by the definition of K, and it carries no analytic content about solutions. Consequently, despite its geometric packaging, it cannot by itself yield a double-sided bound on u(x_2,t_2)/u(x_1,t_1).

minor comments (4)

[Throughout] The manuscript contains numerous typographical errors, including "dimentional," "T wo," "W arsa w," and "hea t equa tion" in the title and header; a careful proofreading pass is needed.
[Sections 2 and 3] The notation for the two intersection points is inconsistent: x* and x* are used where x_* and x^* are clearly intended; this ambiguity should be resolved for readability.
[Section 2.2] The sentence "Since the plane is parallel to the vertical axis, it means γ = 0" is confusing; presumably the plane is vertical in the sense that it contains the t-direction, so its normal has zero t-component. The wording should be clarified.
[Figures] Figures 1 and 2 are referenced in the text but are not embedded in the manuscript version provided; the authors should ensure that the figures appear in the final submission.

Circularity Check

1 steps flagged · score 8.0 of 10

Theorem 3.1 is a definitional identity: the chord ratio and the kernel-ratio expression are the same product of Euclidean distances by construction, so the advertised Harnack–geometry connection is a renaming rather than a derivation.

self definitional [Section 3, Theorem 3.1, eq. (3.1); definitions of K and D_ab in Sections 2.4 and 3]
"Then the following formula holds D20D13/D10D23 = ( K(A,x*)/K(B,x*) K(B,x*)/K(A,x*) )^{1/(n+1)}, where Dab are the chords described by the expressions D20 = sqrt(t_B^2+|B'-x*|^2), D13 = sqrt(t_A^2+|A'-x*|^2), D10 = sqrt(t_A^2+|A'-x*|^2), D23 = sqrt(t_B^2+|B'-x*|^2)."

The Poisson kernel is defined as K(x,t;y)=c_n t/(t^2+|x-y|^2)^{(n+1)/2}. Hence K(A,x*)/K(B,x*) = (t_A/t_B)(D20^2/D10^2)^{(n+1)/2} and K(B,x*)/K(A,x*) = (t_B/t_A)(D13^2/D23^2)^{(n+1)/2}. Multiplying these two ratios and taking the 1/(n+1) power returns exactly D20D13/(D10D23). The theorem therefore follows solely by substituting the definitions of K and D_ab; no property of positive solutions u, no Widder representation, and no Harnack estimate is used. The claimed connection is a notational identity, forced by construction.

full rationale

The only n-dimensional theorem, Theorem 3.1, is an algebraic identity: both sides are the same product of Euclidean distances from A and B to x* and x*, rewritten using the definition of the Poisson kernel. It contains no solution u and no double-sided Harnack inequality. The abstract advertises a higher-dimensional connection to a sharp Harnack bound, but Section 4 explicitly frames the higher-dimensional circle as a 'conjecture' and 'postulate' and does not supply the Widder representation theorem or the kernel-extremal estimates needed to pass from kernel ratios to u(x2,t2)/u(x1,t1). The self-citation to [2] supplies the 1D Harnack theorem and is not what forces (3.1); its reuse is not itself circular. However, since the paper calls (3.1) 'the central point of the present work' and that central point reduces by definition to a tautology, the circularity score is high under the self-definitional pattern. The missing n-dimensional Harnack bound is a correctness/completeness concern rather than an additional circular step.

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

The central identity rests on standard geometry and on the cited Poisson kernel representation. The main missing datum is the n-dimensional Widder representation, which is not proved here.

assumptions (3)

standard math Euclidean geometry: for two space-time points A,B in R^n x R_+ there exists a circle in their vertical plane with center on t=0 and intersections x*,x*.
Invoked in Sections 2 and 3 to define x*,x*; no proof is given in the n-dimensional case.
domain assumption The Poisson kernel K(x,t;y)=c_n t/(t^2+|x-y|^2)^((n+1)/2) is the fundamental solution of the fractional heat equation with half-Laplacian.
Cited to reference [3]; used to convert distance ratios into kernel ratios in (2.8) and Theorem 3.1.
domain assumption Positive classical solutions are represented by Widder convolution with the Poisson kernel.
Imported from references [1,2] for the 1D case; the present paper does not prove the n-dimensional version, but the intended Harnack connection requires it.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Geometric aspects of the Harnack Inequality for a nonlocal heat equation." pith.science (2026). https://pith.science/paper/NBJQ3ZA4

@misc{pith2026250608187,
  author       = {Pith},
  title        = {Pith review of: Geometric aspects of the Harnack Inequality for a nonlocal heat equation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NBJQ3ZA4}},
  note         = {Machine review of arXiv:2506.08187}
}

read the original abstract

We establish a connection between a sharp double-sided Harnack bound for positive solutions of a fractional heat equation and the circular geometry in higher dimensions. The present work extends and generalizes the results obtained in the preceding paper.

Figures

Figures reproduced from arXiv: 2506.08187 by the authors.

**Figure 1.** Semicircle 2. Geometrical approach 2.1. Two dimentional inspiration. We begin our discussion with an observation concerning distance in the hyperbolic metric in the Poincaré halfspace R × R+. Given a semicircle centered on the horizontal axis, the hyperbolic distance, that is, the arc length between two points A and B is expressed by the absolute value of log( d20d13 d10d23 ), where dab denote the corresponding … view at source ↗

**Figure 2.** Semicircle in the special plane 2.3. Circle radius and center. To determine the center and radius of the desired circle, we employ well-known formulas that lead to systems of equations. The plane is defined by the equation (x2 − x1)(y − y1) − (y2 − y1)(x − x1) = 0. (2.3) [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

8 extracted references · 8 canonical work pages

[2]

A sharp Harnack bound for a nonlocal heat equation

M. Dembny and M. Sierżęga , A sharp harnack bound for a nonlocal heat equation, arXiv:2303.08186, (2023)

work page Pith review arXiv 2023
[1]

Barrios, I

B. Barrios, I. Peral, F. Soria, and E. V aldinoci , A Widder’s type theorem for the heat equation with nonlocal diffusion, Arch. Ration. Mech. Anal., 213 (2014), pp. 629–650

work page 2014
[3]

Garof alo, Fractional thoughts, in New developments in the analysis of nonlocal operators, vol

N. Garof alo, Fractional thoughts, in New developments in the analysis of nonlocal operators, vol. 723 of Contemp. Math., Amer. Math. Soc., [Providence], RI, [2019] ©2019, pp. 1–135

work page 2019
[4]

Hadamard , Extension à l’équation de la chaleur d’un théorème de A

J. Hadamard , Extension à l’équation de la chaleur d’un théorème de A. Harnack, Rend. Circ. Mat. Palermo (2), 3 (1954), pp. 337–346 (1955). 16 M. DEMBNY AND M. SIERŻĘGA

work page 1954
[5]

R. S. Hamilton , Li-Yau estimates and their Harnack inequalities, in Geometry and analysis. No. 1, vol. 17 of Adv. Lect. Math. (ALM), Int. Press, Somerville, MA, 2011, pp. 329–362

work page 2011
[6]

Li and S.-T

P. Li and S.-T. Yau , On the parabolic kernel of the Schrödinger operator, Acta Math., 156 (1986), pp. 153–201

work page 1986
[7]

Pini , Sulla soluzione generalizzata di Wiener per il primo problema di valori al contorno nel caso parabolico, Rend

B. Pini , Sulla soluzione generalizzata di Wiener per il primo problema di valori al contorno nel caso parabolico, Rend. Sem. Mat. Univ. Padova, 23 (1954), pp. 422–434

work page 1954
[8]

Weber and R

F. Weber and R. Zacher , Li–Yau inequalities for general non-local diffusion equa- tions via reduction to the heat kernel, Math. Ann., 385 (2023), pp. 393–419. F aculty of Mathematics, Mechanics and Informatics, University of W arsa w, ul. Banacha 2, 02-097 W arsa w, Poland Email address: m.dembny@student.uw.edu.pl F aculty of Mathematics, Mechanics and I...

work page 2023

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [2]

A sharp Harnack bound for a nonlocal heat equation

M. Dembny and M. Sierżęga , A sharp harnack bound for a nonlocal heat equation, arXiv:2303.08186, (2023)

work page Pith review arXiv 2023

[2] [1]

Barrios, I

B. Barrios, I. Peral, F. Soria, and E. V aldinoci , A Widder’s type theorem for the heat equation with nonlocal diffusion, Arch. Ration. Mech. Anal., 213 (2014), pp. 629–650

work page 2014

[3] [3]

Garof alo, Fractional thoughts, in New developments in the analysis of nonlocal operators, vol

N. Garof alo, Fractional thoughts, in New developments in the analysis of nonlocal operators, vol. 723 of Contemp. Math., Amer. Math. Soc., [Providence], RI, [2019] ©2019, pp. 1–135

work page 2019

[4] [4]

Hadamard , Extension à l’équation de la chaleur d’un théorème de A

J. Hadamard , Extension à l’équation de la chaleur d’un théorème de A. Harnack, Rend. Circ. Mat. Palermo (2), 3 (1954), pp. 337–346 (1955). 16 M. DEMBNY AND M. SIERŻĘGA

work page 1954

[5] [5]

R. S. Hamilton , Li-Yau estimates and their Harnack inequalities, in Geometry and analysis. No. 1, vol. 17 of Adv. Lect. Math. (ALM), Int. Press, Somerville, MA, 2011, pp. 329–362

work page 2011

[6] [6]

Li and S.-T

P. Li and S.-T. Yau , On the parabolic kernel of the Schrödinger operator, Acta Math., 156 (1986), pp. 153–201

work page 1986

[7] [7]

Pini , Sulla soluzione generalizzata di Wiener per il primo problema di valori al contorno nel caso parabolico, Rend

B. Pini , Sulla soluzione generalizzata di Wiener per il primo problema di valori al contorno nel caso parabolico, Rend. Sem. Mat. Univ. Padova, 23 (1954), pp. 422–434

work page 1954

[8] [8]

Weber and R

F. Weber and R. Zacher , Li–Yau inequalities for general non-local diffusion equa- tions via reduction to the heat kernel, Math. Ann., 385 (2023), pp. 393–419. F aculty of Mathematics, Mechanics and Informatics, University of W arsa w, ul. Banacha 2, 02-097 W arsa w, Poland Email address: m.dembny@student.uw.edu.pl F aculty of Mathematics, Mechanics and I...

work page 2023