Pith. sign in

REVIEW 3 major objections 5 minor 23 references

On Strong Markushevich bases $\{t^{\lambda_n}\}_{n=1}^{\infty}$ in their closed span in $L^2 (0, 1)$ and characterizing a subspace of $H^2 (\mathbb{D})$

T0 review · 3 major / 5 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 power-function systems with sparse real exponents form strong Markushevich bases in their closed span in $L^2(0,1)$, yielding compact non-normal operators with spectral synthesis and a characterization of a subspace…

desk verdict New strong Markushevich basis result for separated Müntz systems, mostly sound, but Lemma 2.2 needs a small missing convergence step. read the letter →

arxiv 2505.24761 v1 pith:O5G3HAD3 submitted 2025-05-30 math.FA math.CV

Elias Zikkos This is my paper
classification math.FAmath.CV MSC 30B6030B5047A10
keywords strongMarkushevichbasisbiorthogonalsystemsMüntzspacesspectralsynthesiscompactoperatorsHardyspaceH^2(D)Müntz-Szásztheoremclosedspan
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 studies the closed span $\overline{M_\Lambda}$ of the power functions $\{t^{\lambda_n}\}$ in $L^2(0,1)$, where the exponents are strictly increasing, have positive gaps, and satisfy $\sum 1/\lambda_n<\infty$. Under these conditions the system is shown to be a strong Markushevich basis: no matter how the indices are split between the original functions and their unique biorthogonal partners $r_n$, the mixed family still spans $\overline{M_\Lambda}$. Every function in the closed span has a series representation $f(x)=\sum \langle f,r_n\rangle x^{\lambda_n}$ that converges uniformly on compact subsets of a slit disk. From this the paper builds a class of compact, non-normal operators on $\overline{M_\Lambda}$, including the scaling operator $T_\rho(f)=f(\rho x)$, that admit spectral synthesis. It also proves that, for integer exponents, membership in the classical $H^2(\mathbb{D},\Lambda)$ space is equivalent to lying in $\overline{M_\Lambda}$ with square-summable biorthogonal coefficients.

What carries the argument

The load-bearing object is the strong Markushevich basis, meaning a basis whose dual system can replace any selection of basis vectors without shrinking the space. The construction uses the unique biorthogonal family $r_n$, with the norm bound $\|r_n\|\le m_\varepsilon(1+\varepsilon)^{\lambda_n}$, and the analytic-continuation phenomenon on the slit disk $\mathbb{D}\setminus(-1,0)$, which lets each $f\in\overline{M_\Lambda}$ be written as $f(z)=\sum\langle f,r_n\rangle z^{\lambda_n}$. The coefficient identification $c_n=\langle f,r_n\rangle$ is what turns the analytic representation into a tool for proving that every mixed system $\{e_n:n\in N_1\}\cup\{r_n:n\in N_2\}$ spans $\overline{M_\Lambda}$. The connection to spectral synthesis is then supplied by the theorem that a compact operator with simple non-zero eigenvalues and trivial kernel admits spectral synthesis exactly when its eigenvector system is hereditarily complete.

What would settle it

Take a sequence with the stated gap and sum conditions, for instance $\Lambda=\{n^2\}$, and compute the biorthogonal family $r_n$ by successive least-squares projections in $L^2(0,1)$. If for some $f\in\overline{M_\Lambda}$ the coefficient $\langle f,r_n\rangle$ differs from the coefficient of $z^{\lambda_n}$ in the analytic extension of $f$ on a slit disk, Theorem 1.1(I) fails. Alternatively, test hereditary completeness directly by checking whether the closed span of odd-indexed $t^{\lambda_n}$ together with even-indexed $r_n$ equals the whole closed span; a smaller span would disprove the strong Markushevich property.

Watch

Extended reading notes

Core claim

The central claim is that the real-power system $\{t^{\lambda_n}\}$ is hereditarily complete in its closed span: for every partition $\mathbb{N}=N_1\cup N_2$, the closed span of $\{t^{\lambda_n}:n\in N_1\}\cup\{r_n:n\in N_2\}$ in $L^2(0,1)$ equals $\overline{M_\Lambda}$. The proof establishes the coefficient identity $c_n=\langle f,r_n\rangle$ for the analytic extension $f(z)=\sum c_n z^{\lambda_n}$, giving a complete series representation on the slit disk. Combining hereditary completeness with a classical theorem relating spectral synthesis to hereditary completeness yields compact, non-normal operators $T(f)=\sum \langle f,r_n\rangle u_n x^{\lambda_n}$ with $|u_n|\le\rho^{\lambda_n}$ that admit spectral synthesis; the scaling operators $T_\rho(f)=f(\rho x)$ are a special case. Finally, when the exponents are integers, $f\in H^2(\mathbb{D},\Lambda)$ holds exactly when $f\in\overline{M_\Lambda}$ and $\sum |\langle f,r_n\rangle|^2<\infty$.

Load-bearing premise

The argument assumes that every function in the closed span of the power functions extends to an analytic function on a slit disk with a power-series representation in the same exponents; if that analytic continuation failed, the biorthogonal coefficients would no longer reproduce the function and the basis and spectral-synthesis conclusions would not follow.

Editorial extensions

If this is right

  • For any partition of the natural numbers, the mixed system $\{t^{\lambda_n}:n\in N_1\}\cup\{r_n:n\in N_2\}$ spans $\overline{M_\Lambda}$, so the basis property is inherited by every subsystem choice.
  • Every $f\in\overline{M_\Lambda}$ is determined by its biorthogonal coefficients $\langle f,r_n\rangle$, which appear as the exponents in a uniformly convergent series on compact subsets of the slit disk.
  • The operators $T_\rho(f)=f(\rho x)$ on $\overline{M_\Lambda}$ are compact and non-normal yet admit spectral synthesis, and the same holds for every operator $T(f)=\sum\langle f,r_n\rangle u_n x^{\lambda_n}$ with $|u_n|\le\rho^{\lambda_n}$.
  • For integer exponent sets, a function in the classical $H^2(\mathbb{D},\Lambda)$ space is exactly a function in $\overline{M_\Lambda}$ whose biorthogonal coefficients are square-summable.
  • The class of operators constructed provides examples where spectral synthesis holds despite non-normality, complementing earlier results showing compactness alone does not guarantee it.

Reading between the lines

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

  • Inference: The characterization of $H^2(\mathbb{D},\Lambda)$ suggests a computational route for testing sparse analytic functions: project onto $\overline{M_\Lambda}$ and check the $\ell^2$ norm of the biorthogonal coefficients; this may be easier than checking Fourier support directly.
  • Inference: The same strong-Markushevich argument might extend to weighted $L^2(0,1)$ spaces or to systems of the form $t^{\lambda_n}$ with complex exponents, whenever an analytic-continuation phenomenon similar to the one used here holds.
  • Inference: Since $T_\rho$ is a natural scaling operator, its invariant subspaces could be studied explicitly through the biorthogonal coefficients, potentially giving concrete invariant-subspace examples beyond the normal case.
  • Inference: The sharpness of the norm bound $\|r_n\|\le m_\varepsilon(1+\varepsilon)^{\lambda_n}$ has not been tested computationally; verifying whether the bound is asymptotically attained could indicate how well-conditioned the power-function system is for numerical approximation.
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.

Referee Report

3 major / 5 minor

Summary. The paper studies the closed span M_Λ of the Müntz system {t^{λ_n}}_{n=1}^∞ in L^2(0,1) under the assumptions Σ 1/λ_n < ∞ and inf(λ_{n+1}-λ_n)>0. It proves that {t^{λ_n}} is a strong Markushevich basis for M_Λ, derives the series representation f(z)=Σ ⟨f,r_n⟩ z^{λ_n} for f∈M_Λ, constructs compact non-normal operators T on M_Λ with point spectrum {u_n} that admit spectral synthesis, and characterizes the Hardy subspace H^2(D,Λ) by the conditions f∈M_Λ and Σ|⟨f,r_n⟩|^2<∞.

Significance. If the identified gaps are filled, the results are significant. The strong Markushevich (hereditarily complete) basis result for general Müntz systems satisfying (1.1) is new and connects to spectral synthesis through Markus's theorem. The construction of a class of compact non-normal operators with spectral synthesis, including the concrete example T_ρ f = f(ρx), is a nice application. The characterization of H^2(D,Λ) in terms of the biorthogonal coefficients is clean and potentially useful. The paper is clearly written and provides a self-contained proof of the biorthogonal family construction.

major comments (3)
  1. [Section 2, Lemma 2.2] The proof asserts 'Hence' after establishing lim_{ρ→1-} ∫_0^1 |f(ρt)|^2 dt = ∫_0^1 |f(t)|^2 dt, but this by itself does not imply ∫_0^1 |f(ρt)-f(t)|^2 dt → 0. One must add that f(ρt)→f(t) pointwise for t∈[0,1) (which follows from the uniform convergence of the Müntz series on compact subintervals), and then apply a standard argument (e.g., Fatou's lemma or the Radon–Riesz property) to conclude L^2 convergence. This step is load-bearing: it is used in Theorem 1.1(I) to identify the coefficients c_n = ⟨f,r_n⟩ and in Theorem 1.1(IV) to prove the strong Markushevich property.
  2. [Section 3.1, Theorem 1.1(I) and (IV)] Lemma 2.2 is applied to the deleted system {t^{λ_n}: n≠k} and to the subsequence system {t^{λ_n}: n∈N1}. The paper does not explicitly note that these subsystems still satisfy condition (1.1), which is required for Lemma 2.2 to apply. While this is immediate from the hypotheses, the applications should be made explicit, especially in Theorem 1.1(IV) where N1 is an arbitrary infinite set.
  3. [Section 3.2, Theorem 1.2, compactness proof] The finite-rank operators T_m are defined by T_m(f)(x)=Σ_{n=1}^m ⟨f,r_n⟩ u_n x^n. Since λ_n is generally not an integer, the exponent should be λ_n, not n. As written, T_m(f) is not a truncation of the operator T defined in (1.4), and the subsequent convergence argument does not go through. Correcting the exponent fixes the proof, but the current text is formally incorrect.
minor comments (5)
  1. [Throughout] The same symbol M_Λ is used for both the Müntz system {t^{λ_n}} and its closed span in L^2(0,1); this is confusing and should be changed (e.g., use E_Λ for the system and M_Λ for the closed span).
  2. [Lemma 2.1] The expression |f(te^{iθ})|^2 appears as |f(te^{iθ|^2} with a missing closing parenthesis; this should be corrected.
  3. [Lemma 2.2] The assertion that the series f(ρt)=Σ c_n ρ^{λ_n} t^{λ_n} converges uniformly on [0,1] is true but should be justified briefly, for instance by noting that the original series converges uniformly on the compact interval [0,ρ] and then composing with u=ρt.
  4. [Lemma 2.3] The inner product is written as ⟨f,g⟩ := ∫_0^1 f(t)g(t) dt, omitting the complex conjugate on the second argument; the definition given in the Introduction (with the conjugate) should be used consistently throughout.
  5. [General typesetting] There are several typographical/OCR artifacts (e.g., 'M¨ untz', 'P∞', 'R 1 0') that should be cleaned up in the final typeset version.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central strong-Markushevich-basis and spectral-synthesis claims are proved from external Müntz-space theorems and self-contained lemmas; the sole self-citation is a proof-scheme remark and is not load-bearing.

full rationale

The derivation chain is self-contained relative to external Müntz-space facts. Theorem 1.1(I) is obtained by combining the Clarkson-Erdős-Schwartz phenomenon (cited to [10], [14, Cor 6.2.4], [8, Thm 6.4]) with the coefficient identification c_n = <f,r_n>, which is proved in the paper through Lemma 2.2 and biorthogonality; no part of that identification assumes the conclusion. The strong Markushevich basis assertion (IV) is then proved by taking f orthogonal to the mixed system and using (1.2) plus Lemma 2.2 to force f = 0; this is a direct argument, not a restatement of the definition. Lemma 2.3 obtains the biorthogonal norm bound from the external Luxemburg-Korevaar distance estimate [17, relation (1.9)]; no parameter is fitted to the desired conclusion. The only self-citation, Remark 1.3 and the opening of Section 3.2, refers to [22] as a proof scheme; the relevant arguments are written out here and the theorem's validity does not depend on accepting [22] as a premise. Theorem 1.3 is a straightforward consequence of (1.2) and uniqueness of power-series coefficients, not a renaming of the input. One non-circular gap exists: in Lemma 2.2 the passage from convergence of the integrals of |f(ρt)|^2 to convergence of |f(ρt)-f(t)|^2 is asserted without proof; it is true under the Lemma's uniform-convergence hypothesis (pointwise a.e. convergence plus norm convergence), so this is an omitted justification rather than circularity. No fitted-input-called-prediction, imported uniqueness, or ansatz-smuggling step was found.

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

The paper introduces no free parameters, no fitted constants, and no new postulated entities. Its load-bearing elements are a set of established theorems from Müntz space theory and spectral synthesis, plus the biorthogonal functions r_n which are constructed, not invented.

assumptions (5)
  • standard math Müntz-Szász theorem: under Σ1/λ_n < ∞ and inf(λ_{n+1}-λ_n)>0, the system {t^{λ_n}} is minimal and its closed span is proper in L^2(0,1).
    Invoked in the Introduction to set up the biorthogonal family and the frame of the problem.
  • standard math Clarkson-Erdős-Schwartz phenomenon: every f in M_Λ extends analytically to D^* and has the series representation (1.2).
    Core input for Theorem 1.1(I) and all subsequent coefficient identification; cited from [10], [14], [8].
  • standard math Luxemburg-Korevaar lower bound for distances in Müntz systems, D_{n,a} ≥ m_ϵ(1-ϵ)^{λ_n}.
    Used in Lemma 2.3 to get the norm bound (2.2) for the biorthogonal family, which is needed in the proof of Theorem 1.2.
  • standard math Markus's Theorem A: compact operator with trivial kernel and simple eigenvalues admits spectral synthesis iff its eigenvector system is hereditarily complete.
    Bridges hereditary completeness of M_Λ to spectral synthesis of the constructed operators; cited from [18, Theorem 4.1].
  • standard math Fatou's lemma and standard Hilbert-space weak/strong convergence facts.
    Used in Lemma 2.2 to prove that a series representation on [0,1) implies membership in M_Λ.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On Strong Markushevich bases $\{t^{\lambda_n}\}_{n=1}^{\infty}$ in their closed span in $L^2 (0, 1)$ and characterizing a subspace of $H^2 (\mathbb{D})$." pith.science (2026). https://pith.science/paper/O5G3HAD3

@misc{pith2026250524761,
  author       = {Pith},
  title        = {Pith review of: On Strong Markushevich bases $\t^\lambda_n\_n=1^\infty$ in their closed span in $L^2 (0, 1)$ and characterizing a subspace of $H^2 (\mathbbD)$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O5G3HAD3}},
  note         = {Machine review of arXiv:2505.24761}
}
abstract

Let $\Lambda=\{\lambda_n\}_{n=1}^{\infty}$ be a strictly increasing sequence of positive real numbers such that $\sum_{n=1}^{\infty}\frac{1}{\lambda_n}<\infty$ and $\inf(\lambda_{n+1}-\lambda_n)>0$. We investigate properties of the closed span of the system $\{t^{\lambda_n}\}_{n=1}^{\infty}$ in $L^2 (0,1)$, denoted by $\overline{M_{\Lambda}}$, and of the unique biorthogonal family $\{r_n (t)\}_{n=1}^{\infty}$ to the system $\{t^{\lambda_n}\}_{n=1}^{\infty}$ in $\overline{M_{\Lambda}}$. We show that the system $\{t^{\lambda_n}\}_{n=1}^{\infty}$ is a strong Markushevich basis in $\overline{M_{\Lambda}}$ and we obtain a series representation for functions in $\overline{M_{\Lambda}}$. We also construct a general class of operators on $\overline{M_{\Lambda}}$ that admit spectral synthesis. In particular, for all $\rho \in (0,1)$ the operator $T_{\rho}(f)=f(\rho x)$ on $\overline{M_{\Lambda}}$ admits spectral synthesis. In addition, we characterize a certain subspace of the classical Hardy space $H^2 (\mathbb{D})$. Under the extra assumption that $\Lambda\subset\mathbb{N}$, let $H^2(\mathbb{D}, \Lambda)$ consist of functions $f$ in $H^2(\mathbb{D})$ so that the Fourier coefficients $c_n$ of the boundary function $f(e^{i\theta})$ vanish for all $n\notin \Lambda$. We prove that $f\in H^2(\mathbb{D}, \Lambda)$ if and only if $f\in\overline{M_{\Lambda}}$ and $\sum_{n=1}^{\infty}\left| \langle f, r_n\rangle \right|^2<\infty$, where $\langle f, g\rangle= \int_{0}^{1} f(t)\cdot \overline{g(t)}\, dt$.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

23 extracted references · 17 canonical work pages

  1. [12]

    Fricain, P

    E. Fricain, P. Lef` evre,L2 - M¨ untz Spaces as Model Spaces, Complex Anal. Oper. Theory13 no. 1 (2019), 127–139

    work page 2019

  2. [22]

    Zikkos, G

    E. Zikkos, G. Gunatillake, Hereditary completeness of exponential systems {eλnt}∞ n=1 in their closed span in L2(a, b) and spectral synthesis, J. Math. Anal. Appl., 541 no. 1 (2025), 128731. 12

    work page 2025

  3. [1]

    Agler, J

    J. Agler, J. E. McCarthy, A generalization of Hardy’s operator and an asymptotic M¨ untz-Sz´ asz theorem, Expo. Math. 40 no. 4 (2022), 920–930

    work page 2022

  4. [2]

    Agler, J

    J. Agler, J. E. McCarthy, Asymptotic M¨ untz-Sz´ asz theorems, Studia Math. 270 no. 3 (2023), 301–322

    work page 2023

  5. [3]

    J. M. Almira, M¨ untz Type Theorems I. Surveys in Approximation Theory 3 (2007), 152–194

    work page 2007

  6. [4]

    Baranov, Y

    A. Baranov, Y. Belov, A. Borichev, Hereditary completeness for systems of exponentials and reproducing kernels, Adv. Math. 235 (2013), 525–554

    work page 2013

  7. [5]

    Baranov, Y

    A. Baranov, Y. Belov, A. Borichev, Spectral Synthesis in de Branges Spaces, Geom. Funct. Anal. 25 (2015), 417–452

    work page 2015

  8. [6]

    Boivin, C

    A. Boivin, C. Zhu, Bi-orthogonal Expansions in the Space L2(0, ∞) in Blaschke products and their applications, Edited by J. Mashreghi and E. Fricain Springer, New York; Fields Institute for Research in Mathematical Sciences, Toronto, ON, 2013. x+319 pp. ISBN:978-1-4614-5340-6 ISBN:978-1-4614-5341-3

    work page 2013

Show all 23 references
  1. [7]

    Borwein, T

    P. Borwein, T. Erd´ elyi, Polynomials and polynomial inequalities, Springer-Verlag, New-York,

  2. [8]

    Borwein, T

    P. Borwein, T. Erd´ elyi, Generalizations of M¨ untz Theorem via a Remez-type inequality for M¨ untz spaces, J. Amer. Math. Soc. 10 (1997), 327–349

    1997

  3. [9]

    Chalendar, E

    I. Chalendar, E. Fricain, D. Timotin, Embedding theorems for M¨ untz spaces, Ann. Inst. Fourier, 61 no. 6 (2011), 2291–2311

    2011

  4. [10]

    J. A. Clarkson, P. Erd˝ os, Approximation by polynomials, Duke Math. J. 10 (1943), 5–11

    1943

  5. [11]

    P. L. Duren, Theory of H p spaces, Pure and Applied Mathematics, 38 (Academic Press, New York, 1970), p. xii+258

    1970

  6. [13]

    Gaillard, P Lef` evre, Lacunary M¨ untz spaces: isomorphisms and Carleson embeddings, Ann

    L. Gaillard, P Lef` evre, Lacunary M¨ untz spaces: isomorphisms and Carleson embeddings, Ann. Inst. Fourier, 68 no. 5 (2018), 2215–2251

    2018

  7. [14]

    V. I. Gurariy, W. Lusky, Geometry of M¨ untz spaces and related questions, Lecture Notes in Mathematics, 1870. Springer-Verlag, Berlin, 2005. xiv+172 pp. ISBN: 978-3-540-28800-8; 3-540- 28800-7

    2005

  8. [15]

    Jevti´ c D

    M. Jevti´ c D. Vukoti´ c, M. Arsenovi´ c, Taylor coefficients and coefficient multipliers of Hardy and Bergman-type spaces, Springer, Electronic ISBN: 978-3-319-45644-7

  9. [16]

    Koosis, Introduction to Hp spaces, Cambridge University Press, Online ISBN: 9780511470950, 2009

    P. Koosis, Introduction to Hp spaces, Cambridge University Press, Online ISBN: 9780511470950, 2009

    2009

  10. [17]

    W. A. J. Luxemburg, J. Korevaar, Entire functions and M¨ untz-Sz´ asz type approximation, Trans. Amer. Math. Soc. 157 (1971), 23–37

    1971

  11. [18]

    A. S. Markus, The problem of spectral synthesis for operators with point spectrum, Math. USSR Izv. 4 no. 3 (1970), 670–696

    1970

  12. [19]

    Mashreghi, Representation theorems in Hardy spaces, Cambridge University Press, Cambridge, 2009, ISBN: 978-0-521-73201-7

    J. Mashreghi, Representation theorems in Hardy spaces, Cambridge University Press, Cambridge, 2009, ISBN: 978-0-521-73201-7

    2009

  13. [20]

    S. W. Noor, D. Timotin, Embeddings of M¨ untz spaces: the hilbertian case, Proc. Amer. Math. Soc. 141 no. 6 (2013), 2009–2023

    2013

  14. [21]

    Wermer, On invariant subspaces of normal operators, Proc

    J. Wermer, On invariant subspaces of normal operators, Proc. Amer. Math. 3 no. 2 (1952), 270–277

    1952

  15. [1995]

    ISBN: 0-387-94509-1

    x+480 pp. ISBN: 0-387-94509-1. 11

Pith tools

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