Pith. sign in

REVIEW 4 minor 4 cited by

Odd-base Cantor measures cannot have Fourier frames

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · glm-5.2

2026-07-10 03:27 UTC pith:HYJEB7G2

load-bearing objection Resolves Strichartz's question: odd-base Cantor measures admit no Fourier frames. Clean proof, machine-checked in Lean.

arxiv 2607.08656 v1 pith:HYJEB7G2 submitted 2026-07-09 math.FA

Cantor measures with odd base do not admit Fourier frames

Jaume de Dios Pont , Lukas Liehr , Mitchell A. Taylor This is my paper
classification math.FA
keywords cantorfourieradmitbaseframemeasureanswersexistence
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper proves that the Cantor measure with base b—an odd integer greater than 1—does not admit a Fourier frame. A Fourier frame is a collection of complex exponentials that provides stable, invertible reconstruction for every function in the L² space of the measure, relaxing the stricter requirement of an orthonormal basis. The result closes a question posed by Strichartz in 2000 about the middle-third Cantor measure (the case b = 3). Combined with the known construction of Jorgensen and Pedersen showing that even-base Cantor measures are spectral (they admit orthonormal Fourier bases), this yields a complete classification: a Cantor measure with integer base b admits a Fourier frame if and only if b is even. The even/odd dichotomy, previously known only for orthonormal bases, thus persists at the level of frames.

Core claim

The central mechanism is a scale-contradiction argument. The authors construct a sequence of trigonometric polynomials p_n on the torus, built from digit sets of the Cantor measure, whose L² norms are normalized to 1. The lower frame inequality forces a uniform positive lower bound on a certain sum Σ X_n(λ) over all frame frequencies λ, independent of the scale n. Simultaneously, a pointwise estimate—enabled by the key inequality cos²(x) ≥ (1/b²) sin²(x) cos²(bx), which holds for odd b—shows that each summand X_n(λ) is dominated by F(λ) = |μ̂_b(λ)|², and the upper frame inequality applied to the constant function gives Σ F(λ) ≤ B. For each fixed λ, the summand X_n(λ) → 0 as n → ∞ because it携

What carries the argument

The proof constructs a measure σ_n on the torus from the frame frequencies and shows it satisfies frame-like inequalities for certain trigonometric polynomials. A Cauchy-Schwarz step upgrades a lower bound involving cos²(πz)|p_n|² to one involving cos⁴(πz)|p_n|². The pointwise bound X_n(λ) ≤ F(λ) then allows dominated convergence on the countable frequency set Λ, forcing the sum to zero while the frame inequality forces it to stay above a positive constant.

Load-bearing premise

The load-bearing step is the elementary inequality cos²(x) ≥ (1/b²) sin²(x) cos²(bx) for odd integers b > 1, stated without proof. This inequality produces the pointwise domination X_n(λ) ≤ F(λ) that enables the dominated convergence argument. If it failed for some x, the domination bound would collapse.

What would settle it

If one could exhibit a Fourier frame for the middle-third Cantor measure (b = 3), the theorem would be false. More locally, if the inequality cos²(x) ≥ (1/b²) sin²(x) cos²(bx) failed for some x with b odd, the pointwise bound X_n(λ) ≤ F(λ) would break and the dominated convergence argument would not yield a contradiction.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 4 minor

Summary. The paper proves that the Cantor measure μ_b with odd integer base b > 1 does not admit a Fourier frame, resolving a question of Strichartz. Combined with the Jorgensen–Pedersen construction for even bases, this yields a complete even/odd classification. The proof proceeds by contradiction: assuming a Fourier frame E(Λ) exists, the authors transfer frame bounds to measures σ_n on the torus T (Proposition 3.6), construct trigonometric polynomials p_n yielding a uniform lower bound on Σ X_n(λ) (equation (4)), and then show via a pointwise domination estimate X_n(λ) ≤ F(λ) (equation (5)) and dominated convergence that the same sums tend to zero. A Lean 4 formalization of the main result is also provided.

Significance. The result cleanly settles a well-known open problem in the spectral theory of fractal measures. The even/odd dichotomy for Fourier frames on Cantor measures is a natural and complete classification that will be of broad interest. The proof is short, self-contained, and elegant in its use of the interplay between the self-similarity of the Fourier transform (Lemma 3.1), a Cauchy–Schwarz step producing the cos^4 lower bound (equation (3)), and the dominated convergence contradiction. The inclusion of a machine-checked Lean 4 formalization is a significant strength that provides independent verification of the logical chain.

minor comments (4)
  1. Lemma 3.2 (the inequality cos²(x) ≥ (1/b²) sin²(x) cos²(bx) for odd b) is stated without proof. While the inequality is elementary and correct, it is load-bearing for the pointwise estimate in equation (5) and hence for the dominated convergence argument. A brief proof (or a reference) should be included for completeness.
  2. §5: The paper states that the Lean formalization verifies the main result, but the reader cannot independently confirm from the manuscript alone whether the formalization is complete (i.e., contains no 'sorry' or 'admit'). A brief remark confirming the absence of sorries, or a link to continuous integration output, would strengthen the formalization claim.
  3. §1.5: The notation Γ_n(τ_n) and the sets Ξ_m appear only in the narrative about the failed construction and play no role in the actual proof. This is fine for context, but a brief sentence clarifying that these objects are not used in the proof would help the reader.
  4. The definition of Q_m in §3 uses the phrase 'with no non-zero digit in positions b^m, b^{m+1}, ...', which is slightly informal. Stating that Q_m consists of non-negative integers less than b^m whose base-b digits all lie in {0,1} would be more explicit for the reader.

Circularity Check

0 steps flagged

No circularity detected: self-contained proof with independent Lean verification

full rationale

The paper's derivation chain is self-contained and non-circular. The main result (Theorem 1.1) is proved by contradiction: assuming a Fourier frame E(Λ) exists for μ_b with odd b, the authors derive two incompatible conclusions — a uniform lower bound Σ X_n(λ) ≥ A²/(2b²B) (equation 4) and convergence Σ X_n(λ) → 0 (via dominated convergence using equations 5 and 6). Each step is derived from first principles or standard results: Lemma 3.1 (the self-similarity F(bx) = cos²(2πx)F(x)) is attributed to Jorgensen-Pedersen [22], an independent external source with no author overlap. Lemma 3.2 (the elementary inequality cos²(x) ≥ (1/b²)sin²(x)cos²(bx) for odd b) is stated without proof but is a standard trigonometric fact, not a self-citation. Propositions 3.5–3.6 and 4.1 are proved in full within the paper. No definition is circularly constructed in terms of the quantity it claims to derive. No parameter is fitted to data and then presented as a prediction. No load-bearing step reduces to a self-citation by the present authors. The Lean 4 formalization provides additional machine-checked verification independent of the prose argument. The proof is genuinely self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

No invented entities. This is pure mathematics.

axioms (3)
  • standard math F(bx) = cos²(2πx)F(x) for the squared modulus of the Fourier transform of μ_b (Lemma 3.1)
    Cited from Jorgensen-Pedersen [22, Section 7]. Used in Proposition 4.1 to derive the product formula for F(λ). Standard result in the theory of self-similar measures.
  • standard math cos²(x) ≥ (1/b²) sin²(x) cos²(bx) for odd integer b > 1 (Lemma 3.2)
    Stated in §3 without proof. Used in the proof of Theorem 1.1 to derive the pointwise estimate X_n(λ) ≤ F(λ) (equation 5), which is essential for the dominated convergence argument. Elementary but load-bearing.
  • standard math Existence and uniqueness of the invariant measure μ_b for the contractive IFS {T_0, T_1} (§2.1)
    Attributed to Hutchinson [20]. Standard result in fractal geometry.

pith-pipeline@v1.1.0-glm · 15153 in / 3928 out tokens · 268944 ms · 2026-07-10T03:27:31.721059+00:00 · methodology

0 comments
read the original abstract

We prove that the Cantor measure with base $b$ does not admit a Fourier frame whenever $b > 1$ is an odd integer. In particular, this answers a question of Strichartz on the existence of a Fourier frame for the middle third Cantor measure. A formalization of our main result in Lean 4 is also provided.

discussion (0)

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

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Walsh-Quotient Obstruction for Fourier Frames on Odd Reciprocal-Power Bernoulli Convolutions

    math.FA 2026-07 accept novelty 7.5

    If ρ^{-m}=B with B odd ≥3 and ρ<1/2, then L² of the symmetric Bernoulli convolution μ_{ρ,d} admits no Fourier frame of exponentials.

  2. A Walsh-Quotient Obstruction for Fourier Frames on Odd Reciprocal-Power Bernoulli Convolutions

    math.FA 2026-07 unverdicted novelty 7.0

    Odd reciprocal-power Bernoulli convolutions admit no Fourier frames in L² when the contraction ratio is the reciprocal of an odd integer power.

  3. A Formalization of the Mean-Field Derivation of the Vlasov Equation

    cs.AI 2026-07 unverdicted novelty 6.0

    AI-assisted Lean 4 formalization yields an axiom-clean, sorry-free development of Dobrushin mean-field well-posedness for the nonlinear Vlasov equation plus a Mathlib-absorbable Wasserstein-1 layer.

  4. A Formalization of the Mean-Field Derivation of the Vlasov Equation

    cs.AI 2026-07 conditional novelty 6.0

    A mathematician directing an AI completed an axiom-clean Lean 4 formalization of Dobrushin's mean-field derivation of the Vlasov equation, including well-posedness, stability, a conditional mean-field limit, and a sho...

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages · cited by 2 Pith papers · 1 internal anchor

  1. [1]

    On spectral Cantor–Moran measures and a variant of Bourgain’s sum of sine problem

    Li-Xiang An, Xiaoye Fu, and Chun-Kit Lai. On spectral Cantor–Moran measures and a variant of Bourgain’s sum of sine problem. Adv. Math. , 349:84–124, 2019

    work page 2019

  2. [2]

    When does a Bernoulli convolution admit a spectrum? Adv

    Xin-Rong Dai. When does a Bernoulli convolution admit a spectrum? Adv. Math. , 231(3– 4):1681–1693, 2012

    work page 2012

  3. [3]

    Spectra of Cantor measures

    Xin-Rong Dai. Spectra of Cantor measures. Math. Ann. , 366(3–4):1621–1647, 2016

    work page 2016

  4. [4]

    Spectral property of Cantor measures with consecutive digits

    Xin-Rong Dai, Xing-Gang He, and Chun-Kit Lai. Spectral property of Cantor measures with consecutive digits. Adv. Math. , 242:187–208, 2013

    work page 2013

  5. [5]

    On spectral N -Bernoulli measures

    Xin-Rong Dai, Xing-Gang He, and Ka-Sing Lau. On spectral N -Bernoulli measures. Adv. Math., 259:511–531, 2014

    work page 2014

  6. [6]

    The Lean 4 Theorem Prover and Programming Language

    Leonardo de Moura and Sebastian Ullrich. The Lean 4 Theorem Prover and Programming Language. In Automated Deduction – CADE 28: 28th International Conference on Au- tomated Deduction, Virtual Event, July 12–15, 2021, Proceedings , page 625–635, Berlin, Heidelberg, 2021. Springer-Verlag

    work page 2021

  7. [7]

    Dutkay, Deguang Han, Qiyu Sun, and Eric Weber

    Dorin E. Dutkay, Deguang Han, Qiyu Sun, and Eric Weber. On the Beurling dimension of exponential frames. Adv. Math. , 226(1):285–297, 2011

    work page 2011

  8. [8]

    Dutkay, John Haussermann, and Chun-Kit Lai

    Dorin E. Dutkay, John Haussermann, and Chun-Kit Lai. Hadamard triples generate self- affine spectral measures. Trans. Amer. Math. Soc. , 371(2):1439–1481, 2019

    work page 2019

  9. [9]

    Dutkay and Chun-Kit Lai

    Dorin E. Dutkay and Chun-Kit Lai. Uniformity of measures with Fourier frames. Adv. Math., 252:684–707, 2014

    work page 2014

  10. [10]

    Existence and exactness of ex- ponential Riesz sequences and frames for fractal measures

    Dorin Ervin Dutkay, Shahram Emami, and Chun-Kit Lai. Existence and exactness of ex- ponential Riesz sequences and frames for fractal measures. J. Anal. Math. , 143(1):289–311, 2021

    work page 2021

  11. [11]

    On the spectra of a Cantor measure

    Dorin Ervin Dutkay, Deguang Han, and Qiyu Sun. On the spectra of a Cantor measure. Adv. Math. , 221(1):251–276, 2009

    work page 2009

  12. [12]

    Continuous and discrete Fourier frames for fractal measures

    Dorin Ervin Dutkay, Deguang Han, and Eric Weber. Continuous and discrete Fourier frames for fractal measures. Trans. Amer. Math. Soc. , 366(3):1213–1235, 2014

    work page 2014

  13. [13]

    Fourier bases and Fourier frames on self-affine measures

    Dorin Ervin Dutkay, Chun-Kit Lai, and Yang Wang. Fourier bases and Fourier frames on self-affine measures. In Julien Barral and Stéphane Seuret, editors, Recent Developments in Fractals and Related Fields , Trends in Mathematics, pages 87–111. Birkhäuser, Cham, 2017

    work page 2017

  14. [14]

    Weighted Fourier frames on fractal measures

    Dorin Ervin Dutkay and Rajitha Ranasinghe. Weighted Fourier frames on fractal measures. J. Math. Anal. Appl. , 444(2):1603–1625, 2016

    work page 2016

  15. [15]

    Translational absolute continuity and Fourier frames on a sum of singular measures

    Xiaoye Fu and Chun-Kit Lai. Translational absolute continuity and Fourier frames on a sum of singular measures. J. Funct. Anal. , 274(9):2477–2498, 2018

    work page 2018

  16. [16]

    Commuting self-adjoint partial differential operators and a group theoretic problem

    Bent Fuglede. Commuting self-adjoint partial differential operators and a group theoretic problem. J. Funct. Anal. , 16(1):101–121, 1974. CANTOR MEASURES WITH ODD BASE DO NOT ADMIT FOURIER FRAMES 15

    work page 1974

  17. [17]

    Strichartz

    Zijian Guo, Rachel Kogan, Hua Qiu, and Robert S. Strichartz. Boundary value problems for a family of domains in the Sierpinski gasket. Illinois J. Math. , 58(2):497–519, 2014

    work page 2014

  18. [18]

    Exponential spectra in L2(µ)

    Xing-Gang He, Chun-Kit Lai, and Ka-Sing Lau. Exponential spectra in L2(µ). Appl. Com- put. Harmon. Anal. , 34(3):327–338, 2013

    work page 2013

  19. [19]

    Spectral property of the Bernoulli convolutions

    Tian-You Hu and Ka-Sing Lau. Spectral property of the Bernoulli convolutions. Adv. Math., 219(2):554–567, 2008

    work page 2008

  20. [20]

    Hutchinson

    John E. Hutchinson. Fractals and self similarity. Indiana Univ. Math. J. , 30(5):713–747, 1981

    work page 1981

  21. [21]

    Fourier frames for surface- carried measures

    Alex Iosevich, Chun-Kit Lai, Bochen Liu, and Emmett Wyman. Fourier frames for surface- carried measures. Int. Math. Res. Not. IMRN , 2022(3):1644–1665, 2022

    work page 2022

  22. [22]

    Palle E. T. Jorgensen and Steen Pedersen. Dense analytic subspaces in fractal L2-spaces. J. Anal. Math. , 75:185–228, 1998

    work page 1998

  23. [23]

    Singular vectors on fractals and projections of self-similar measures

    Osama Khalil. Singular vectors on fractals and projections of self-similar measures. Geom. Funct. Anal., 30(2):482–535, 2020

    work page 2020

  24. [24]

    On spectral Cantor measures

    Izabella Łaba and Yang Wang. On spectral Cantor measures. J. Funct. Anal. , 193(2):409– 420, 2002

    work page 2002

  25. [25]

    On Fourier frame of absolutely continuous measures

    Chun-Kit Lai. On Fourier frame of absolutely continuous measures. J. Funct. Anal. , 261(10):2877–2889, 2011

    work page 2011

  26. [26]

    Non-spectral fractal measures with Fourier frames

    Chun-Kit Lai and Yang Wang. Non-spectral fractal measures with Fourier frames. J. Fractal Geom., 4(3):305–327, 2017

    work page 2017

  27. [27]

    Fourier frames for singular measures and pure type phenomena

    Nir Lev. Fourier frames for singular measures and pure type phenomena. Proc. Amer. Math. Soc., 146(7):2883–2896, 2018

    work page 2018

  28. [28]

    Fourier Frames on Salem Measures

    Longhui Li and Bochen Liu. Fourier frames on Salem measures. arXiv:2506.01280, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  29. [29]

    Exponential frames on un- bounded sets

    Shahaf Nitzan, Alexander Olevskii, and Alexander Ulanovskii. Exponential frames on un- bounded sets. Proc. Amer. Math. Soc. , 144(1):109–118, 2016

    work page 2016

  30. [30]

    Gabriel Picioroaga and Eric S. Weber. Fourier frames for the Cantor-4 set. J. Fourier Anal. Appl., 23(2):324–343, 2017

    work page 2017

  31. [31]

    dense analytic subspaces in fractal L2-spaces

    Robert S. Strichartz. Remarks on “dense analytic subspaces in fractal L2-spaces” by P. E. T. Jorgensen and S. Pedersen. J. Anal. Math. , 75:229–231, 1998

    work page 1998

  32. [32]

    Strichartz

    Robert S. Strichartz. Mock Fourier series and transforms associated with certain Cantor measures. J. Anal. Math. , 81:209–238, 2000

    work page 2000

  33. [33]

    The Lean Mathematical Library

    The mathlib Community. The Lean Mathematical Library. In Proceedings of the 9th ACM SIGPLAN International Conference on Certified Programs and Proofs , CPP 2020, New Orleans, LA, USA, January 2020. ACM. Center for Data Science, New York University, New York, New York 10011, USA Email address : jdedios@nyu.edu Department of Mathematics, Bar-Ilan University...

    work page 2020