Uniform local Lipschitz continuity of eigenvalues with respect to the potential in $L^1[a,b]$

Jiangang Qi; Xiao Chen

REVIEW 2 major objections 3 minor 16 references

Uniform local Lipschitz continuity of eigenvalues with respect to the potential in $L^1[a,b]$

T0 review · 2 major / 3 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 for regular Sturm-Liouville problems with a monotone, strictly positive weight, the entire eigenvalue sequence $\lambda_n(q)$ is uniformly locally Lipschitz in the $L^1$ potential: every bounded set $\Omega$ admits…

desk verdict Plausible uniform Lipschitz bound for eigenvalues in L1, but the key Lemma 3.2 is not proved: (3.22) fails even for constant weight and constant g. read the letter →

arxiv 1908.05527 v1 pith:H5V2YNXI submitted 2019-08-15 math.CA

Xiao Chen , Jiangang Qi This is my paper · ORCID

classification math.CA MSC 34B0545J0534L15

keywords Sturm-LiouvilleproblemeigenvalueuniformlocalLipschitzcontinuityL1potentialPrüfertransformationmonotoneweightnormalizedeigenfunctionsboundedvariation

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 new uniformity property for the eigenvalues of a regular Sturm-Liouville problem: once the potential $q$ is confined to a bounded subset of $L^1([0,1],\mathbb{R})$, a single constant bounds how much every eigenvalue can move when $q$ changes. The main theorem assumes the weight $\omega$ is monotone and satisfies $\inf \omega > 0$, and concludes $|\lambda_n(q_1)-\lambda_n(q_2)| \le C(\Omega)\|q_1-q_2\|_{L^1}$ for every $n$ and every $q_1,q_2$ in the bounded set $\Omega$. This matters because individual eigenvalues are known to be differentiable with derivative given by the square of the normalized eigenfunction; the new work supplies the missing uniform-in-$n$ control of those eigenfunctions, which is what turns differentiability into a uniform Lipschitz estimate.

What carries the argument

The central object is the elliptic Pr\'ufer transformation: a solution $y$ is written through an amplitude $\rho(x;\lambda)$ and angle $\theta(x;\lambda)$, with $y = \rho \sin\theta/\sqrt{\lambda}$ and $y' = \rho \cos\theta$. The amplitude's growth is governed by $H(x;\lambda) = (\sqrt{\lambda}/2)\int_0^x (1-\omega(t)) \sin 2\theta(t;\lambda)\,dt$, which appears as the exponent in $\rho(x;\lambda) = \rho(0;\lambda)\exp(H(x;\lambda))$. Lemma 3.2 is the load-bearing estimate: for a monotone weight bounded below and any function $g$ of bounded variation, the integrals of $g(x)\sin 2\theta(x;\lambda)$ and $g(x)\cos 2\theta(x;\lambda)$ are $O(1/\sqrt{\lambda})$ uniformly. This estimate makes $H(x;\lambda)$ uniformly bounded, which in turn bounds the Pr\'ufer amplitudes of normalized eigenfunctions independently of $n$; that uniform bound converts the derivative formula into the uniform local Lipschitz conclusion.

What would settle it

A decisive check is to work through the last interval $[x_m, \bar{x}]$ in the proof of Lemma 3.2 with explicit monotone functions $g$ and $\omega$: verify whether the asserted bound in equation (3.22) follows from the monotonicity sandwich in (3.15)-(3.16) when $\theta(\bar{x};\lambda)$ lies between $m\pi+\pi/2$ and $(m+1)\pi$. Finding one choice of $g$, $\omega$, and $\lambda$ for which the claimed upper bound fails would disprove the lemma; alternatively, a repaired inequality for that interval would confirm the main theorem's foundation.

Watch

Extended reading notes

Core claim

On its own terms, Theorem 2.1 is the claim: for the eigenvalue problem $-y''+q(x)y = \lambda\omega(x)y$ on $[0,1]$ with separated self-adjoint boundary conditions, if the weight $\omega$ satisfies H1 (monotonicity) and H2 ($\inf \omega > 0$), then the eigenvalue sequence $\{\lambda_n(q)\}$ is uniformly locally Lipschitz continuous with respect to $q$ in $L^1([0,1],\mathbb{R})$. The proof uses the Fr\'echet derivative formula $\partial \lambda_n/\partial q \cdot h = \int_0^1 \phi_n^2 h$, where $\phi_n$ is the normalized eigenfunction, and reduces the difficulty to Proposition 3.5, which states that on any bounded set of potentials the normalized eigenfunctions are uniformly bounded in $x$ and $n$. With that bound, integrating the derivative along a line segment from $q_1$ to $q_2$ gives the Lipschitz estimate directly. The uniform eigenfunction bound is obtained through elliptic Pr\'ufer coordinates and a Riemann-Lebesgue-type estimate for oscillatory integrals against the Pr\'ufer angle.

Load-bearing premise

Everything rests on Lemma 3.2's claim that bounded-variation functions have oscillatory integrals against $\sin$ and $\cos$ of the Pr\'ufer angle decaying uniformly like $O(1/\sqrt{\lambda})$; the proof of that lemma, in the passage around equation (3.22), does not correctly justify the bound on the final partial oscillation interval, and without that justification the uniform eigenfunction bound and the main theorem are not established.

Editorial extensions

If this is right

Bounded $L^1$ sets of potentials have a uniform modulus of continuity for the whole spectrum, so no single eigenvalue can be singled out for worse behavior.
Approximating a potential in $L^1$ by a sequence automatically gives quantitative convergence of every eigenvalue with the same rate constant.
The result transfers through the Liouville transformation to problems with general coefficient $p$, since the transformation preserves eigenvalues and provides an equivalent $L^1$ metric on the transformed potential.
The bound is independent of $n$, so spectral computations over a bounded potential family can use a constant $C(\Omega)$ without tracking eigenvalue index.

Reading between the lines

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

Not stated in the paper: the monotonicity hypothesis H1 may be stronger than necessary; a natural next test is whether a non-monotone weight with $\inf \omega > 0$ already admits uniform eigenfunction bounds or admits a counterexample.
Not stated in the paper: the same oscillatory-integral technique could yield uniform local Lipschitz estimates for eigenvalue gaps or for matrix Sturm-Liouville systems, where Pr\'ufer-type transformations exist.
Not stated in the paper: if the uniform eigenfunction bound holds, the Lipschitz estimate could be differentiated along curves in $L^1$ to give stronger differentiability information, not just continuity.

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.

Desk Editor's Note

read the letter

The short version: this paper proves something worth knowing — a uniform (n-independent) local Lipschitz bound for Sturm-Liouville eigenvalues with respect to the potential in L1, under monotone weights bounded away from zero. That's genuinely new and would be a useful tool. The proof strategy via the Prüfer angle and oscillatory-integral estimates is sensible, and the paper is clearly written. The authors correctly rely on known differentiability of eigenvalues and do not smuggle in the conclusion. Credit where due: the organization is honest and the main idea is natural.

The problem is Lemma 3.2. The stress-test note is right: inequality (3.22) is false in the simplest case ω≡1, g≡1. Then f(t) is the constant 1, but the partial integral over the last oscillation interval can be 1/2 > 0. So the claimed upper bound of 0 fails. This is not a small typo; the lemma is used to bound the Prüfer amplitude H(x;λ) in Lemma 3.3, which underpins the uniform eigenfunction bound in Proposition 3.5, which is the load-bearing step for Theorem 2.1. Without a valid proof of Lemma 3.2, the main theorem is currently unproved.

The likely reality is that the lemma's statement is true and the gap is fixable — the estimate is the kind of thing that follows from a more careful second mean value theorem or a sharper treatment of the last interval. But the manuscript does not supply that proof. I'd flag this to the authors as a major revision, not a rejection: they need to repair Lemma 3.2 before the paper can stand.

Who gets value from this paper? People working in inverse spectral theory or spectral optimization who want quantitative stability of eigenvalues. For them, the result matters if it holds, and the current gap is an obstacle.

My recommendation: send it to peer review, but with the expectation that the referee will catch the broken inequality. The paper deserves a serious referee because the result is plausible, the novelty is real, and the flaw is specific and addressable rather than fatal to the approach. Don't desk-reject; but don't accept without a corrected Lemma 3.2.

Referee Report

2 major / 3 minor

Summary. The manuscript claims the following theorem: for the regular Sturm-Liouville problem (2.4)-(2.5) on [0,1] with p≡1 and weight ω satisfying H1 (monotonicity) and H2 (inf ω > 0), the sequence of eigenvalues {λ_n(q)} is uniformly locally Lipschitz in q with respect to the L1 norm, with a constant independent of n on each bounded subset of L1. The proof proceeds in three steps: (i) an oscillatory-integral estimate (Lemma 3.2) for ∫ g sin2θ and ∫ g cos2θ, used to bound the Prüfer amplitude uniformly in λ; (ii) a uniform bound on normalized eigenfunctions on L1-bounded sets of potentials (Proposition 3.5), proved via a Volterra-Gronwall argument; (iii) an application of the derivative formula for eigenvalues to integrate the eigenfunction bound along a line segment between two potentials (Theorem 2.1).

Significance. If the theorem is correct, it is a meaningful strengthening of the known continuity and differentiability results for Sturm-Liouville eigenvalues, and the statement is clean and falsifiable. The approach is analytic and self-contained apart from standard cited facts; I see no circularity or parameter fitting. The main obstacle is the proof of Lemma 3.2: the key estimate is not established as written, and the proof of Proposition 3.5 also omits a boundary-condition case. These are substantial but local gaps, and the announced result remains plausible.

major comments (2)

[§3, Lemma 3.2, Eq. (3.22)] The last-interval estimate (3.22) is false as stated. For ω(x)≡1 and g(x)≡1, one has f≡1, θ(x;λ)=√λ x (with θ(0)=0), and for θ(x̄)=mπ+3π/4 the partial integral on [x_m,x̄] equals ∫_{mπ}^{mπ+3π/4} sin 2u du = 1/2 > 0, while the asserted upper bound f(x_m)-f(x̄) is 0. Since (3.22) is used in (3.24) to control the last interval and hence to obtain the O(1) bound on G(c;λ) in (3.27), Lemma 3.2 --- and through it Lemma 3.3 and Proposition 3.5 --- is not proved as written. The lemma may be true with a different final-interval bound, but the present proof does not supply one.
[§3, Proposition 3.5, Eq. (3.37) and (3.50)] The proof assumes 'We may as well assume that C1 ≠ 0' after (3.37), but this excludes the Dirichlet boundary condition at the left endpoint: when α=0, the boundary condition y(0) cosα + y'(0) sinα = 0 forces y(0)=C1=0 for every nontrivial eigenfunction. The subsequent estimates (3.50)-(3.57) divide by C1^2 and therefore do not apply in that case. Since Proposition 3.5 is stated for all α,β∈[0,π), the proof needs a separate treatment of the case C1=0, for example by taking the φ-solution as the leading term.

minor comments (3)

[§3, Eq. (3.26)] The first integral in the display is written with 'dt' although the integration variable is x; this should be corrected to dx.
[Throughout] There are repeated typos and misspellings, including 'Lesbegue', 'Prüfrer', 'Riemann-Lesbegue', 'syetems', 'Trnas.', and 'Probelm'; these should be cleaned up before publication.
[§3, Lemma 3.4] The proof of Lemma 3.4 is too terse: the passage from absolutely continuous g to arbitrary L1 g, and the passage from weights ω+1/n to ω, require a fuller justification because the stated continuity of the Prüfer angle in the weight is not obviously uniform in λ. For Proposition 3.5, this issue can be avoided by applying Lemma 3.2 directly to the bounded-variation weight ω, since H2 is in force there.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Theorem 2.1 is proved from standard external theorems via an analytic Prüfer argument; no input is renamed as a prediction.

full rationale

The paper's central claim, uniform local Lipschitz continuity of the eigenvalue sequence λ_n(q), is derived by combining the Fréchet derivative formula cited from Kong and Zettl (Theorem 2.2) with a uniform bound on normalized eigenfunctions (Proposition 3.5), obtained through Prüfer transformation, Gronwall estimates, and the auxiliary Lemma 3.2. The hypotheses H1 and H2 on the weight function do not contain the Lipschitz conclusion, and the proof proceeds by a parameter-dependent path q_t = q_1 + tΔq and the mean-value representation |λ_n(q_2)-λ_n(q_1)| ≤ ∫∫ φ_n^2 |Δq|, which is bounded using Proposition 3.5. No parameter is fitted to the target quantity, and no 'prediction' is constructed from data or from the desired inequality itself. The citations involving an author of the present paper ([11], [16]) appear only as background or for a standard Liouville transformation and are not load-bearing for Theorem 2.1; the load-bearing cited results are Theorem 2.2 from Kong-Zettl and standard Prüfer theory from Zettl's monograph, which are external and independent. Even if the last-interval estimate (3.22) in Lemma 3.2 contains a genuine proof gap — a correctness concern, not a circularity concern — the argument does not reduce to its inputs by definition, by fitted parameters, or by a self-citation chain. The derivation is therefore self-contained in the sense relevant to circularity.

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

No free parameters are fitted; all constants are explicit bounds from the proof. The only non-standard inputs are the hypotheses H1 and H2 on the weight. The proof relies on standard tools from Sturm-Liouville theory and integral inequalities.

assumptions (5)

domain assumption H1: ω is monotonic on [0,1]
Assumed in Theorem 2.1 and used in Lemma 3.2 to control the sign of sin2θ and to bound the sums over oscillation intervals.
domain assumption H2: inf_{x∈[0,1]} ω(x) > 0
Assumed in Theorem 2.1; used to make the auxiliary function f(t) finite and to bound denominators in Lemma 3.2. Also used in Proposition 3.5.
standard math Standard spectral theory: the regular Sturm-Liouville problem has discrete real eigenvalues bounded below
Used in the introduction and in Proposition 3.5 via λ_n→∞.
standard math Differentiability of eigenvalues (Kong-Zettl Theorem 2.2)
Cited as [3, Theorem 4.2(6)]; used in the proof of Theorem 2.1 to express the derivative as ∫ φ_n^2 h.
standard math Gronwall inequality
Used in Proposition 3.5 to bound the solution y(x;λ) of the initial value problem.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Uniform local Lipschitz continuity of eigenvalues with respect to the potential in $L^1[a,b]$." pith.science (2026). https://pith.science/paper/H5V2YNXI

@misc{pith2026190805527,
  author       = {Pith},
  title        = {Pith review of: Uniform local Lipschitz continuity of eigenvalues with respect to the potential in $L^1[a,b]$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H5V2YNXI}},
  note         = {Machine review of arXiv:1908.05527}
}

abstract

The present paper shows that the eigenvalue sequence $\{\lambda_n(q)\}_{n\geqslant 1}$ of regular Sturm-Liouville eigenvalue problem with certain monotonic weights is uniformly Lipschitz continuous with respect to the potential $q$ on any bounded subset of $L^1([a,b],\mathbb{R})$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[1]

J. H. Barrett, A Pr¨ ufer transformation for matrix diﬀerentia l equations, Proc. Amer. Math. Soc. , 8(1957) 510–518

work page 1957
[2]

J. Chu, G. Meng and M. Zhang, Continuity and minimization of spect rum related with the periodic Camassa-Holm equation, J. Diﬀ. Equ. , 265(4) (2018) 1678–1695

work page 2018
[3]

Kong and A

Q. Kong and A. Zettl, Eigenvalues of regular Sturm-Liouville problem , J. Diﬀ. Equ. , 131 (1996) 1–19

work page 1996
[4]

Meng and M

G. Meng and M. Zhang, Measure diﬀerential equations II: Contin uity of eigenvalues in measures with weak ∗ topology, technical report, 2009

work page 2009
[5]

Meng and M

G. Meng and M. Zhang, Continuity in weak topology: First order line ar syetems of ODE, Acta China Sinica Engl. Ser. , 26 (2010) 1287–1298

work page 2010
[6]

Moeller and A

M. Moeller and A. Zettl, Diﬀerentiable dependence of eigenvalues o f operators in Banach spaces, J. Operator. Theory., 36 (1996) 335–355

work page 1996
[7]

B. G. Pachpatte, Inequalities for diﬀerential and integral equations , Academic Press, San Diego, 1998

work page 1998
[8]

P¨oschel and E

J. P¨oschel and E. Trubowitz, Inverse spectral theory , Pure. Appl Math., vol. 130, Academic Press, New York, 1987

work page 1987

Show all 16 references

[9]

Zettl, Sturm-Liouville theory , Math

A. Zettl, Sturm-Liouville theory , Math. Surveys Monogr., vol. 121, Amer. Math. Soc., Providence. RI, 2005

2005
[10]

W. T. Reid, A Pr¨ ufer transformation for diﬀerential systems , Paciﬁc J. Math. , 8(3) (1958) 575–584

1958
[11]

Xie and J

B. Xie and J. Qi, No-real eigenvalues of indeﬁnite Sturm-Liouville p robelm, J. Diﬀ. Equ. , 255(1958) 2291–2301

1958
[12]

Z. Wen, M. Yang and M. Zhang, Complete continuity of eigen-pair s of weighted Dirichlet eigenvalue problem, Mediterr J. Math. , 15:73 (2018)

2018
[13]

Yan and M

P. Yan and M. Zhang, Continuity in weak topology and extremal p roblems of eigenvalues of the p-laplacian, Trnas. Amer. Math. Soc. , 363 (2011) 2003–2028

2011
[14]

Zhang, Continuity in weak topology: High order linear syetems of ODE, Sci

M. Zhang, Continuity in weak topology: High order linear syetems of ODE, Sci. China Ser. A , 51 (2008) 1038–1058

2008
[15]

Zhang, A review on strong continuity results of solutions and eigenvalues in inﬁnitely dimensional parameters, technical report, 2014

M. Zhang, A review on strong continuity results of solutions and eigenvalues in inﬁnitely dimensional parameters, technical report, 2014

2014
[16]

Zhang, Z

M. Zhang, Z. Wen, G. Meng, J. Qi and B. Xie, On the number and c omplete continuity of weighted eigenvalues of measure diﬀerential equations, Diﬀ. Integ. Equ. , 31(9-10) (2018) 761–784. School of Mathematics and Statistics, Shandong University at Weihai, Weihai 264209, P.R. Ch...

2018

Pith tools

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

[1] [1]

J. H. Barrett, A Pr¨ ufer transformation for matrix diﬀerentia l equations, Proc. Amer. Math. Soc. , 8(1957) 510–518

work page 1957

[2] [2]

J. Chu, G. Meng and M. Zhang, Continuity and minimization of spect rum related with the periodic Camassa-Holm equation, J. Diﬀ. Equ. , 265(4) (2018) 1678–1695

work page 2018

[3] [3]

Kong and A

Q. Kong and A. Zettl, Eigenvalues of regular Sturm-Liouville problem , J. Diﬀ. Equ. , 131 (1996) 1–19

work page 1996

[4] [4]

Meng and M

G. Meng and M. Zhang, Measure diﬀerential equations II: Contin uity of eigenvalues in measures with weak ∗ topology, technical report, 2009

work page 2009

[5] [5]

Meng and M

G. Meng and M. Zhang, Continuity in weak topology: First order line ar syetems of ODE, Acta China Sinica Engl. Ser. , 26 (2010) 1287–1298

work page 2010

[6] [6]

Moeller and A

M. Moeller and A. Zettl, Diﬀerentiable dependence of eigenvalues o f operators in Banach spaces, J. Operator. Theory., 36 (1996) 335–355

work page 1996

[7] [7]

B. G. Pachpatte, Inequalities for diﬀerential and integral equations , Academic Press, San Diego, 1998

work page 1998

[8] [8]

P¨oschel and E

J. P¨oschel and E. Trubowitz, Inverse spectral theory , Pure. Appl Math., vol. 130, Academic Press, New York, 1987

work page 1987