The strong log-concavity for first eigenfunction of the Ornstein-Uhlenbeck operator in the class of convex bodies

Lei Qin

REVIEW 4 major objections 6 minor 20 references

The strong log-concavity for first eigenfunction of the Ornstein-Uhlenbeck operator in the class of convex bodies

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · 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 for every open bounded convex domain, the negative logarithm of the first positive Dirichlet eigenfunction of the Ornstein-Uhlenbeck operator is strictly convex, and uses this to characterize equality in the Gaussian…

desk verdict A plausible improvement of strong log-concavity to general convex domains, but the proof as written is conditional on unstated hypotheses of the cited constant-rank theorem. read the letter →

arxiv 2507.00819 v1 pith:TTPDLG3F submitted 2025-07-01 math.AP

Lei Qin This is my paper · ORCID

classification math.AP MSC 35J2535P1552A4026B25

keywords stronglog-concavityfirsteigenvalueOrnstein-UhlenbeckoperatorconvexbodiesBrunn-MinkowskiinequalityconstantranktheoremGaussianprincipalfrequency

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

The paper's central claim is that the negative logarithm of the first positive Dirichlet eigenfunction of the Ornstein-Uhlenbeck operator is strictly convex on every open bounded convex domain, with no smoothness or symmetry assumptions on the boundary. This upgrades an earlier log-concavity result to strong log-concavity: the Hessian matrix of $w=-\ln u$ is positive definite at every interior point, so the eigenfunction has no flat or ridge directions in the Gaussian log-scale. As a consequence, the paper also characterizes the equality case in the Brunn-Minkowski inequality for the Gaussian principal frequency: if equality holds for some convex combination of two bounded convex domains, the two domains coincide up to translation. The proof proceeds through a constant-rank theorem for the Hessian of $w$, showing that its rank cannot drop, and then rules out any rank below $n$ by the boundary blow-up $w\to+\infty$ at $\partial\Omega$.

What carries the argument

The central object is the Hessian matrix $D^2w$ of $w=-\ln u$, and the central mechanism is the constant-rank theorem for this Hessian. The proof studies the trace $\varphi(x)=\operatorname{tr}(D^2w)\ge 0$, deriving the differential inequality $\Delta\varphi\le c_1|\nabla\varphi|+c_2\varphi$ near a point of minimum rank, and invokes the strong minimum principle to force $\varphi\equiv 0$ locally, so the rank of $D^2w$ is constant throughout $\Omega$. A second ingredient, carried over from constant-rank theory, says that a convex function of constant rank $r<n$ has flat or linear directions; the boundary blow-up $w\to+\infty$ at $\partial\Omega$ rules those out, leaving rank $n$.

What would settle it

Take a rectangular box in $\mathbb{R}^n$, for example a long thin rectangle in $\mathbb{R}^2$, solve the Ornstein-Uhlenbeck eigenvalue problem (1) numerically, and compute the eigenvalues of $D^2(-\ln u)$ at interior points near the boundary and near a flat side; finding a zero or negative eigenvalue at any interior point would disprove Theorem 1.1.

Watch

Extended reading notes

Core claim

Theorem 1.1 states that if $\Omega$ is an open, bounded, convex subset of $\mathbb{R}^n$ and $u>0$ is a solution of the Ornstein-Uhlenbeck Dirichlet problem (1), then $w=-\ln u$ is strongly convex, meaning $D^2w(x)>0$ for every $x\in\Omega$. The discovery is that the extra hypotheses used in the earlier result, namely strictly positively curved $C^{2,\alpha}$ boundary and origin symmetry, are not needed. The structural reason is that the Hessian $D^2w$ has constant rank in $\Omega$: if the rank were some $r<n$, the convex function $w$ would be constant along $n-r$ coordinate directions or linear along at least one direction, which contradicts the boundary condition $w\to+\infty$ as $x\to\partial\Omega$. With constant rank $n$, $D^2w>0$ follows. The paper further proves that equality in the Brunn-Minkowski inequality for the Gaussian principal frequency forces the two domains to coincide up to translation, again without the earlier smoothness and symmetry assumptions.

Load-bearing premise

The load-bearing premise is that the constant-rank theorem for the Hessian of $w$, proved in [6] only under extra smoothness and symmetry assumptions, applies to all bounded convex domains, together with the unproved strong minimum principle for the trace inequality; if either fails, the constant-rank step collapses.

Editorial extensions

If this is right

The first positive eigenfunction of the Ornstein-Uhlenbeck operator is now known to be strongly log-concave on every bounded convex domain, not merely on smooth, origin-symmetric ones.
The equality case of the Brunn-Minkowski inequality for the Gaussian principal frequency holds without the earlier smoothness and symmetry hypotheses: equality for a convex combination forces the two domains to coincide up to translation.
The constant-rank property of the Hessian of $w=-\ln u$ is established on general bounded convex domains, which is a structural fact usable in further variational problems.
Since $\Delta w>0$ follows from $D^2w>0$, the proof recovers the earlier log-concavity theorem as a special case via a different route.

Reading between the lines

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

The same constant-rank-plus-boundary-blow-up strategy should apply to other self-adjoint operators with convexity-preserving structure, such as weighted $p$-Laplacian versions of the Ornstein-Uhlenbeck operator, where log-concavity has already been studied.
If the strong log-concavity theorem is correct, equality in the Gaussian Brunn-Minkowski inequality is rigid in the strong sense that any equality pair must be translates, which suggests that quantitative stability estimates could be derived from the positivity of $D^2w$.
A testable extension is whether the Hessian of $w$ stays uniformly positive with a lower bound depending only on diameter, inradius, and the Gaussian weight; the present proof only gives pointwise positivity.

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

Dear Colleague,

The thing to know: this short paper claims strong log-concavity of the first eigenfunction for the Ornstein–Uhlenbeck operator on every bounded convex domain, removing the C^{2,α}_+ boundary and origin-symmetry assumptions that [6] needed. If true, that is a clean improvement. The proof is a constant-rank argument, and the direct computation for the rank-zero case is basically sound.

The genuinely new content is Theorem 1.1. It is a real step beyond the log-concavity theorem in [6], and the equality-case Theorem 1.2 is a plausible extension. I found no signs of parameter fitting or invented machinery; the author is building on an established program.

The soft spots are about presentation and verification of borrowed results.

First, the decisive step in Claim 1 for minimum rank r between 1 and n is outsourced to [6, Theorem 4.5] without stating that theorem's hypotheses. [6] proves strong log-concavity only under extra assumptions, so the reader cannot tell whether the constant-rank theorem applies to arbitrary bounded convex domains. This is load-bearing. The author needs to state the hypotheses and either cite a version that is known to hold in full generality or prove it. It may well be true, but as written the proof is conditional.

Second, the strong minimum principle used to go from inequality (3) to φ ≡ 0 is asserted without statement or reference. It is standard, but it should be identified. Third, Theorem 1.2 is stated with 'we omit the proof'; for a main theorem, a sketch or a precise pointer to identical steps in [5,6] would be better.

The calculation behind (3) is more or less fine: the quadratic term Σ w_ii^2 is absorbed by φ because w_ii ≥ 0, so no flaw there. The boundary-blow-up contradiction in Claim 2 is schematic but believable.

Who is this for? Convex geometers and people working on Brunn–Minkowski inequalities for eigenvalues. A serious referee should look at it, mainly to verify the applicability of [6, Theorem 4.5] and to ask for the missing statements. I would not cite it in the current form, because the central implication is not self-contained, but I would send it to review.

Best.

Referee Report

4 major / 6 minor

Summary. The paper claims that for any open, bounded, convex domain Ω ⊂ R^n, the function w = −ln u, where u > 0 is the first Dirichlet eigenfunction of the Ornstein–Uhlenbeck operator, is strongly convex in Ω (Theorem 1.1). The proof combines the log-concavity result from [6], a constant-rank argument for the Hessian D²w (with the rank-0 case handled by a differential inequality and a strong minimum principle), and a boundary blow-up argument to rule out rank below n. The paper also states an equality case for the Brunn–Minkowski type inequality for the Gaussian principal frequency (Theorem 1.2), with proof omitted.

Significance. If Theorem 1.1 is correct, it removes the extra assumptions (C^{2,α}_+ boundary and origin symmetry) imposed in [6] for the strong log-concavity of the Gaussian principal frequency eigenfunction, and it would settle a natural question in the constant-rank-theorem approach to log-concavity. The overall strategy is standard and plausible, and the paper is concise. However, the proof is incomplete as written: the decisive constant-rank step for rank r ≥ 1 is outsourced to a theorem from [6] whose hypotheses are not stated or verified, and the strong minimum principle used in the rank-0 case is asserted without proof or reference. These are load-bearing gaps because the final conclusion rests on them.

major comments (4)

[Section 3, Claim 1] The proof for the case 1 ≤ r ≤ n consists of the sentence 'the case has been considered in [6, Theorem 4.5]'. The hypotheses of [6, Theorem 4.5] are not stated, and the paper does not verify that they hold for an arbitrary bounded convex domain. This is particularly concerning because the paper's own Theorem 2.3 shows that the strong results in [6] require ∂Ω ∈ C^{2,α}_+ and origin symmetry; if [6, Theorem 4.5] carries similar restrictions, then Claim 1 is not established for the general setting of Theorem 1.1, and the subsequent contradiction in Claim 2 cannot proceed. The author must either state [6, Theorem 4.5] in full and prove that its hypotheses are satisfied for every bounded convex Ω, or provide a self-contained constant-rank proof.
[Section 3, inequality (3) and strong minimum principle] After deriving the differential inequality Δφ ≤ c1|∇φ| + c2φ, the paper states 'By the strong minimum principle, this implies that φ ≡ 0 in a neighborhood of x0', but no statement, proof, or citation is given. The inequality is not of the classical form for which the standard strong minimum principle applies, and the positive gradient term requires a dedicated argument (for example, via Taylor expansion or unique continuation). Since the conclusion φ ≡ 0 is essential to establish constant rank when r = 0, this lemma should be stated explicitly and proved or cited precisely.
[Section 3, derivation of (3)] The constants in inequality (3) are not tracked correctly. With the stated bounds |z| ≤ c1, |∇w| ≤ c2, and |w_ii| ≤ c2 (with w_ii ≥ 0), the displayed computation yields coefficients (c1 + 2c2) for |∇φ| and (2 + 2c2) for φ, not (2c1 + c2) and (2 + c2) as written. Also, the step bounding 2Σ w_ii² by 2c2φ uses the inequalities w_ii ≥ 0 and w_ii ≤ c2, which should be stated. This is a local imprecision, but since (3) is the mechanism for the r = 0 case, the derivation should be corrected.
[Theorem 1.2] Theorem 1.2 is advertised in the abstract as a characterization of the equality case of the Brunn–Minkowski inequality, improving the corresponding result in [6], but its proof is omitted entirely ('we omit the proof'). Because the hypotheses differ from those in [5,6] (no symmetry or boundary regularity), the reduction is not automatic; the author should provide a detailed sketch or a precise statement of how the arguments in [5,6] adapt to the present situation.

minor comments (6)

[Introduction] The sentence 'we note that as shown in [6], if the origin belongs to the domain and it is the maximum point of u, it is obvious that (x, ∇w(x)) ≥ 0 ... and then, we have Δw > 0 ... following those steps in [3, 14]' is too terse; please expand the chain of implications for readability.
[Introduction] There is a typo: 'which relies onConstant Rank theorem' should read 'which relies on the constant rank theorem'.
[Section 2] The phrase 'open subsect of Rn' is a typo; it should be 'open subset of Rn'.
[Section 2] The symbol ∂Ω is used in the eigenvalue definition before being defined; please define it at first use.
[Section 3] The formula after 'By a direct calculation' contains unmatched parentheses in the term Σ_i (|∇w|²)_ii; please re-typeset for clarity.
[References] The citation [6] is to an arXiv preprint; the theorem numbers quoted from it (e.g., Theorem 4.5, Proposition 4.5) should be checked for consistency with the numbering in the published or latest version of [6].

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the proof is a theorem-deduction chain that relies on external results from [6], not on fitted inputs or self-citation.

full rationale

The paper proves Theorem 1.1 by deriving a constant-rank property for D^2w and then ruling out constant rank r < n. The load-bearing invocation is [6, Theorem 4.5] for the case 1 <= r <= n, and [6, Proposition 4.5] with [14] for Claim 2. These are external works by other authors, not the present author's own prior results, so the reliance is ordinary mathematical dependency rather than self-citation. No parameter is fitted to data and then renamed as a prediction; no quantity is defined in terms of the target conclusion; no known result is repackaged under new coordinates. The only self-citations ([7], [18]) appear in the introduction and bibliography as background context and are not used in the proof of Theorem 1.1 or Theorem 1.2. The reviewer's concern that the hypotheses of [6, Theorem 4.5] are not stated and may not cover arbitrary bounded convex domains is a completeness or correctness risk, not circularity: the cited theorem, if applicable, supplies independent support, and if it is not applicable the proof is incomplete rather than circular. Similarly, the unproved strong minimum principle invoked for inequality (3) and the omitted proof of Theorem 1.2 are proof gaps, not circular reductions. The paper does not define its conclusion into existence, and none of its steps is equivalent to its inputs by construction.

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

The central proof rests on the cited log-concavity theorem and constant-rank results from [6], plus an unstated maximum-principle lemma and the standard elliptic regularity of the eigenfunction. No free parameters or new entities are introduced.

assumptions (5)

domain assumption Existence, positivity, and C∞ regularity of the eigenfunction u for bounded Lipschitz domains
Invoked in Section 1 and Section 2, citing [8, 11].
domain assumption The eigenfunction u is log-concave, i.e., w = -ln u is convex on Ω
Theorem 2.1 restates [6, Theorem 1.7] and is used as the starting point in the proof of Theorem 1.1.
domain assumption If the minimum rank of D²w is r ≥ 1, then D²w has constant rank in Ω
Used in Claim 1 of Section 3, attributed to [6, Theorem 4.5] without stating its hypotheses.
domain assumption If D²w has constant rank r < n, then through each point there is a line on which w is affine
Used in Claim 2 of Section 3, citing [6, Proposition 4.5] and [14, pp. 29-31].
standard math A nonnegative C² function φ with an interior zero and satisfying Δφ ≤ c1|∇φ| + c2 φ must vanish in a neighborhood
Invoked as 'the strong minimum principle' in Claim 1, Section 3, but not stated or proved in the paper.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The strong log-concavity for first eigenfunction of the Ornstein-Uhlenbeck operator in the class of convex bodies." pith.science (2026). https://pith.science/paper/TTPDLG3F

@misc{pith2026250700819,
  author       = {Pith},
  title        = {Pith review of: The strong log-concavity for first eigenfunction of the Ornstein-Uhlenbeck operator in the class of convex bodies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TTPDLG3F}},
  note         = {Machine review of arXiv:2507.00819}
}

abstract

In this paper, we prove that the first (positive) Dirichlet eigenvalue of the Ornstein-Uhlenbeck operator \[ L(u)=\Delta u-(\nabla u,x), \] is strongly log-concave if the domain is bounded and convex, which improves the conclusion in [6]. We also provide a characterization of the equality case of the Brunn-Minkowski inequality for the principal frequency of $L(u)$ in the class of convex bodies.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 19 canonical work pages

[6]

Colesanti, E

A. Colesanti, E. Francini, G. Livshyts and P. Salani, The Brunn-Minkowski inequalities for the first eigenvalue of the Ornstein-Uhlenbeck operator and log-concavity of the relevant eigenfunction, arXiv:2407.21354

arXiv
[1]

A. F. Acker, L. E. Payne and G. Philippin, On the convexity of level lines of the fundamental mode in the clamped membrane problem, and the existence of convex solutions in a related free boundary proble, Z. Angew. Math. Phys. 32 (1981), no. 6, 683–694

work page 1981
[2]

H. J. Brascamp and E. H. Lieb, On extensions of the Brunn-Minkowski and Pr ´ekopa-Leindler theorems, including inequalities for log concave functions, and with an application to the diffusion equation , J. Functional Analysis 22 (1976), no. 4, 366–389

work page 1976
[3]

L. A. Caffarelli, A. Friedman, Convexity of solutions to semilinear elliptic equations , Duke Math. J. 52 (2) (1985), pp. 431–456

work page 1985
[4]

Carbotti, S

A. Carbotti, S. Cito, D.A. La Manna, D. Pallara, Stability of the Gaussian Faber-Krahn inequality Ann. Mat. Pura Appl. 203 (2024), 2185-2198

work page 2024
[5]

Colesanti, Brunn-Minkowski inequalities for variational functionals and related problems, Adv

A. Colesanti, Brunn-Minkowski inequalities for variational functionals and related problems, Adv. Math.194 (2005), no. 1, 105–140

work page 2005
[7]

Colesanti, L

A. Colesanti, L. Qin, P. Salani, Geometric properties of solutions to elliptic PDE’s in Gauss space and related Brunn- Minkowski type inequalities, arXiv:2502.00184

work page arXiv
[8]

L. C. Evans, Partial differential equations, second edition, Graduate Studies in Mathematics, 19, Amer. Math. Soc., Providence, RI, 2010. 6 L. QIN

work page 2010

Show all 20 references

[9]

Franceschi, A

V . Franceschi, A. Pinamonti, G. Saracco and G. Stefani, The Cheeger problem in abstract measure spaces, J. Lond. Math. Soc. (2) 109 (2024), no. 1, Paper No. e12840, 55 pp

2024
[10]

Kilpel ¨ainen, Weighted Sobolev spaces and capacity, Ann

T. Kilpel ¨ainen, Weighted Sobolev spaces and capacity, Ann. Acad. Sci. Fenn. Ser. A I Math.19 (1994), no. 1, 95–113

1994
[11]

Gilbarg and N

D. Gilbarg and N. Trudinger, Elliptic partial differential equations of second order , reprint of the 1998 edition. Classics in Mathematics. Springer-Verlag, Berlin, 2001. xiv+517 pp

1998
[12]

Herscovici and G

O. Herscovici and G. V . Livshyts, Kohler-Jobin meets Ehrhard: the sharp lower bound for the Gaussian principal frequency while the Gaussian torsional rigidity is fixed, via rearrangements , Proc. Amer. Math. Soc. 152 (2024), no. 10, 4437–4450

2024
[13]

Kawohl, Rearrangements and convexity of level sets in PDE , Lecture Notes in Mathematics, 1150, Springer, Berlin, 1985

B. Kawohl, Rearrangements and convexity of level sets in PDE , Lecture Notes in Mathematics, 1150, Springer, Berlin, 1985

1985
[14]

Korevaar, J

N. Korevaar, J. Lewis, Convex solutions to certain elliptic equations have constant rank Hessians, Arch. Rat. Mech. Anal. 97 (1987), 19–32

1987
[15]

P. Liu, X. N. Ma and L. Xu, A Brunn-Minkowski inequality for the Hessian eigenvalue in three-dimensional convex domain, Adv. Math. 225 (2010), no. 3, 1616–1633

2010
[16]

X. N. Ma and L. Xu, The convexity of solution of a class Hessian equation in bounded convex domain inR3, J. Funct. Anal. 255 (2008), no. 7, 1713–1723

2008
[17]

N. S. Papageorgiou, V . R˘adulescu and D. Repovˇs, Nonlinear analysis—theory and methods, Springer Monographs in Mathematics, Springer, Cham, 2019

2019
[18]

Qin, Log-concavity of eigenfunction and Brunn-Minkowski inequality of eigenvalue for weighted p-Laplace oper- ator, arXiv:2411.16377

L. Qin, Log-concavity of eigenfunction and Brunn-Minkowski inequality of eigenvalue for weighted p-Laplace oper- ator, arXiv:2411.16377

arXiv
[19]

Salani, A Brunn-Minkowski inequality for the Monge-Amp`ere eigenvalue, Adv

P. Salani, A Brunn-Minkowski inequality for the Monge-Amp`ere eigenvalue, Adv. Math. 194 (2005), no. 1, 67–86

2005
[20]

Salani, Convexity of solutions and Brunn-Minkowski inequalities for Hessian equations in R3, Adv

P. Salani, Convexity of solutions and Brunn-Minkowski inequalities for Hessian equations in R3, Adv. Math. 229 (2012), no. 3, 1924–1948. INSTITUTE OF MATHEMATICS , H UNAN UNIVERSITY , C HANGSHA , 410082, C HINA Email address: qlhnumath@hnu.edu.cn

2012

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [6]

Colesanti, E

A. Colesanti, E. Francini, G. Livshyts and P. Salani, The Brunn-Minkowski inequalities for the first eigenvalue of the Ornstein-Uhlenbeck operator and log-concavity of the relevant eigenfunction, arXiv:2407.21354

arXiv

[2] [1]

A. F. Acker, L. E. Payne and G. Philippin, On the convexity of level lines of the fundamental mode in the clamped membrane problem, and the existence of convex solutions in a related free boundary proble, Z. Angew. Math. Phys. 32 (1981), no. 6, 683–694

work page 1981

[3] [2]

H. J. Brascamp and E. H. Lieb, On extensions of the Brunn-Minkowski and Pr ´ekopa-Leindler theorems, including inequalities for log concave functions, and with an application to the diffusion equation , J. Functional Analysis 22 (1976), no. 4, 366–389

work page 1976

[4] [3]

L. A. Caffarelli, A. Friedman, Convexity of solutions to semilinear elliptic equations , Duke Math. J. 52 (2) (1985), pp. 431–456

work page 1985

[5] [4]

Carbotti, S

A. Carbotti, S. Cito, D.A. La Manna, D. Pallara, Stability of the Gaussian Faber-Krahn inequality Ann. Mat. Pura Appl. 203 (2024), 2185-2198

work page 2024

[6] [5]

Colesanti, Brunn-Minkowski inequalities for variational functionals and related problems, Adv

A. Colesanti, Brunn-Minkowski inequalities for variational functionals and related problems, Adv. Math.194 (2005), no. 1, 105–140

work page 2005

[7] [7]

Colesanti, L

A. Colesanti, L. Qin, P. Salani, Geometric properties of solutions to elliptic PDE’s in Gauss space and related Brunn- Minkowski type inequalities, arXiv:2502.00184

work page arXiv

[8] [8]

L. C. Evans, Partial differential equations, second edition, Graduate Studies in Mathematics, 19, Amer. Math. Soc., Providence, RI, 2010. 6 L. QIN

work page 2010