Pith. sign in

REVIEW 3 major objections 4 minor 21 references

Inverse problem for semi-infinite Jacobi matrices and associated Hilbert spaces of analytic functions

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A semi-infinite Jacobi matrix in the limit circle case determines an infinite-dimensional de Branges space built from its connecting operator.

desk verdict The paper's new semi-infinite de Branges construction and its main theorems rest on a false Chebyshev orthogonality, so the central claims fail as written; the finite-dimensional part and Lemma 1 are sound. read the letter →

arxiv 2505.08338 v1 pith:E5X4LLUP submitted 2025-05-13 math.SP math.CA

Alexander Mikhaylov , Victor Mikhaylov This is my paper · ORCID
classification math.SPmath.CA MSC 47B3644A6046E22
keywords Jacobimatricessemi-infinitedeBrangesspacesboundarycontrolmethodmomentproblemsHankelconnectingoperatorlimitcirclecase
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 tries to extend the authors' boundary-control construction of de Branges spaces from finite Jacobi matrices to semi-infinite ones. In the limit circle (indeterminate) case, where the Jacobi operator has deficiency indices $(1,1)$ and the associated moment problem has many measures, the paper defines a Hilbert space $B_{\infty}^{A}$ of entire functions $F(\lambda)=\sum_{k=1}^{\infty} f_k T_k(\lambda)$ by declaring the formal connecting operator $C$ to provide the norm, $[F,F]_{B_{\infty}^{A}}=C[f,f]$. The claim is that this space has the reproducing kernel $J_z^{\infty}(\lambda)=\sum_{n=1}^{\infty} p_n(z)p_n(\lambda)$ and satisfies the axioms of de Branges space theory, so it is a genuine infinite-dimensional de Branges space attached to the Jacobi matrix. This matters because it gives a dynamical, control-theoretic route to de Branges spaces for semi-infinite Jacobi matrices, in parallel with the classical moment-problem and Hankel-matrix analysis, and it connects the minimal eigenvalues of the finite connecting operators with the limit point/limit circle dichotomy.

What carries the argument

The central object is the connecting operator $C_T$: the Gram matrix of the reachable set $\{u^f_{\cdot,T}\mid f\in F_T\}$ of the discrete dynamical system (2), with entries $\{C_T\}_{ij}=\sum_{k=0}^{T-\max\{i,j\}} r_{|i-j|+2k}$ built from the response vector $(r_0,r_1,\ldots)$; its infinite counterpart $C=(W)^*W$ supplies the norm on $B_{\infty}^{A}$. The bridge to the classical moment problem is the identity $C_T=\Lambda_T S_T \Lambda_T^*$, where $S_T$ is the truncated Hankel matrix and $\Lambda_T$ is the change of basis from monomials to the Chebyshev polynomials of the second kind $T_k(\lambda)$. The proof of Theorem 7 converts the variational formula for $\beta_T$ into a Gram-matrix problem for the polynomials $p_k$, using the claimed orthogonality of the $T_k$ against $d\nu(\lambda)=\chi_{(-1,1)}(\lambda)/\sqrt{1-\lambda^2}\,d\lambda$; Theorem 1, the de Branges classification criterion, is then the test that certifies $B_{\infty}^{A}$ as a de Branges space.

What would settle it

Compute $\int_{-1}^{1} T_1(\lambda)T_3(\lambda)\,\frac{d\lambda}{\sqrt{1-\lambda^2}}$ with $T_1=1$ and $T_3=\lambda^2-1$. The integral equals $-\pi/2$, not $0$, so the claimed orthogonality of the $T_k$ with respect to $d\nu$ is false; consequently the norm identity $\|f\|_{\ell^2}^2=\int |\sum f_kT_k|^2 d\nu$ fails, and the proof of the lower bound (26) in Theorem 7 needs a different justification.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the finite-time construction of de Branges spaces via reachable sets survives the passage $T\to\infty$ exactly in the indeterminacy regime. The main object is the semi-infinite connecting operator $C=(W)^*W$, whose finite truncations $C_T$ are Gram matrices of the reachable set of the discrete wave equation (2). Theorem 7 asserts that in the limit circle case the minimal eigenvalue $\beta_T$ of $C_T$ is bounded below by a positive constant, whose reciprocal is bounded by $\int_{-1}^{1} \ell^{-1}(\lambda)\,d\lambda/\sqrt{1-\lambda^2}$ with $\ell^{-1}(\lambda)=\sum_{k=0}^{\infty}|p_k(\lambda)|^2$; Theorem 8 asserts that the quadratic form $C[f,f]$ is closable in $\ell^2$ in the limit circle case and also for bounded Jacobi matrices with absolutely continuous spectral measure. With closability and the existence of the kernel $J_z^{\infty}(\lambda)=\sum_{n=1}^{\infty} p_n(z)p_n(\lambda)$ in hand, the paper defines $B_{\infty}^{A}$ as the linear manifold of series with $C[f,f]<\infty$ and claims that the two de Branges axioms of Theorem 1 hold, so $B_{\infty}^{A}$ is a de Branges space. The proof of the lower bound in Theorem 7 uses the claimed orthogonality of the $T_k$ with respect to the Chebyshev weight.

Load-bearing premise

The load-bearing premise is that the polynomials $T_k$ defined by $T_{t+1}+T_{t-1}-\lambda T_t=0$, $T_1=1$, are orthogonal with respect to the Chebyshev weight $d\nu(\lambda)=\chi_{(-1,1)}(\lambda)/\sqrt{1-\lambda^2}\,d\lambda$, so that $\|f\|_{\ell^2}^2=\int_{-1}^{1}|\sum f_kT_k(\lambda)|^2 d\nu$; this identity is what converts $\ell^2$ norms of controls into $L^2$ norms in the proof of Theorem 7.

Editorial extensions

If this is right

  • In the limit circle case, the connecting-operator norm gives an intrinsic construction of a de Branges space for a semi-infinite Jacobi matrix, with no need to pass to a canonical system first.
  • The reproducing kernel of this space is the classical kernel $\sum_{n=1}^{\infty}p_n(z)p_n(\lambda)$ from indeterminate moment theory, giving that kernel a dynamical interpretation as the limit of finite-time reproducing kernels $J_z^T(\lambda)=\sum_{n=1}^{T}p_n(z)p_n(\lambda)$.
  • The minimal eigenvalues $\beta_T$ of the finite connecting operators stay bounded away from zero in the limit circle case, the dynamic counterpart of the Berg–Chen–Ismail small-eigenvalue result for Hankel matrices; the paper also proves a converse direction in Lemma 1: boundedness of the maximal eigenvalues $\gamma_T$ forces the limit point case.
  • The quadratic form $C[f,f]$ is closable in the limit circle case and for bounded absolutely continuous spectra, so $B_{\infty}^{A}$ is a genuine Hilbert space and the identification with a de Branges space is legitimate.

Reading between the lines

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

  • Editorial consequence: if the Chebyshev orthogonality used in Theorem 7 fails, the quantitative lower bound on $\beta_T$ needs a different proof; the space $B_{\infty}^{A}$ itself may still be a de Branges space, but the stated eigenvalue estimate and its connection to the classical Hankel result would have to be reworked.
  • Editorial suggestion: a concrete test of the construction is to compute $B_{\infty}^{A}$ for the free Jacobi operator ($a_k=1$, $b_k=0$), where $C_T=I_T$; the resulting space should coincide with the classical de Branges space of the associated canonical system, and identifying its Hermite–Biehler function would settle whether the infinite limit reproduces the known object.
  • Editorial speculation: the same boundary-control recipe could be tested on multidimensional discrete wave equations, where finite speed of propagation should still give finite-dimensional reachable sets; the authors mention this direction in passing.
  • Editorial observation: the paper itself notes that $\beta_T$ alone cannot distinguish the limit point case, so the load-bearing content of the construction lies in the closability of $C[\cdot,\cdot]$ and the existence of the reproducing kernel, not in a sharp spectral criterion from $\beta_T$.
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.

Referee Report

3 major / 4 minor

Summary. The manuscript studies discrete-time dynamical systems associated with finite and semi-infinite Jacobi matrices and proposes to associate special Hilbert spaces of analytic functions, specifically de Branges spaces, with these systems. For finite matrices the construction was developed in the authors' earlier work, and the present paper reviews that construction, connects it with classical moment problems and Hankel matrices, and then attempts to extend it to the semi-infinite (limit circle) case. The main new claims are Theorem 7, a lower bound for the minimal eigenvalue of the finite connecting operators C_T in the indeterminate case; Theorem 8, closability of the quadratic form C[·,·] in the limit circle case and in a bounded case; and the construction of an infinite-dimensional space B∞_A with reproducing kernel J∞_z(λ)=Σ p_n(z)p_n(λ), claimed to be a de Branges space. The paper also contains a useful comparison of properties of the connecting operator matrices and classical Hankel matrices, including the relation C_T=Λ_T S_T Λ_T^* and a lemma linking boundedness of the largest eigenvalue of C_T to the limit point case.

Significance. If the central construction were correct, the paper would give a dynamical, control-theoretic realization of de Branges spaces for semi-infinite Jacobi matrices, going beyond the finite-dimensional case treated in the authors' earlier work, and would establish a new connection between the connecting operators of the boundary control method and the classical moment problem. Some parts of the paper are valuable independently: the review of finite-dimensional results is coherent, the relation C_T=Λ_T S_T Λ_T^* is explicit and useful, and Lemma 1 is a clean and correct argument using the same relation to detect the limit point case. However, the new infinite-dimensional claims rest on a false orthogonality identity for the polynomials T_k, and this error is load-bearing in the proofs of Theorems 7 and 8 and in the construction of B∞_A. As a result, the principal advertised results are not established as written, and the error is internal to the derivation rather than a matter of convention or presentation.

major comments (3)
  1. [Section 6.1, proof of Theorem 7] The proof asserts, immediately before Eq. (28), that the polynomials T_k satisfying T_{t+1}+T_{t-1}-λT_t=0 with T_1=1 are orthogonal with respect to dν(λ)=χ_{(-1,1)}(λ)/√(1-λ²)dλ, so that ∥f∥²=Σ|f_k|²=∫|F(λ)|²dλ/√(1-λ²). This identity is false. For T_1=1 and T_3=λ²-1, ∫_{-1}^{1} T_1(λ)T_3(λ)dλ/√(1-λ²)= -π/2, not 0. The true orthogonality for these polynomials holds on (-2,2) with weight √(4-λ²), not on (-1,1) with the reciprocal square-root weight. Since this identity is used to convert the ℓ² normalization of the control f into the L²(dν) normalization of F and hence to obtain the variational problems (28) and (29), the optimization problem is not equivalent as claimed, and the trace estimate leading to the lower bound (26) is not established.
  2. [Section 6.1, proof of Theorem 8] The closability argument uses the same false identity to pass from f^{(n)}→0 in ℓ² to F^{(n)}→0 in L²(-1,1;dλ/√(1-λ²)). This step is essential: without it, the contradiction argument for the limiting function F in the definition of closability has no premise, and the conclusion that F=0 does not follow. The subsequent line 'F=0 in L²(-1,1;dρ), so F=0 on (-1,1)' also mixes up the measures; the intended measure appears to be dν, not dρ. Even after correcting this notational slip, the key convergence assertion remains false because the polynomials T_k are not orthogonal with respect to dν. Thus the closability of the quadratic form C[·,·] in the limit circle case (Theorem 8(a)) is not proved.
  3. [Section 6.1, definition of B∞_A and the de Branges space claim] The space B∞_A is defined as {Σ f_k T_k(λ) : C[f,f]<∞} with scalar product C[f,g]=∫F(λ)G(λ)dM, and it is asserted that the conditions of Theorem 1 verifying that B∞_A is a de Branges space are 'trivially checked'. This is not a proof. One must show that B∞_A is a Hilbert space of entire functions, that the map from controls f with C[f,f]<∞ to functions is injective and complete with respect to the C-norm, that the integral representation equals the dynamic norm for all such f, and that the reproducing kernel is indeed given by J∞_z(λ)=Σ p_n(z)p_n(λ). These properties depend on the same faulty ℓ²-L²(dν) link and on limiting arguments that are not supplied. Hence the central infinite-dimensional construction, which is the advertised result of the paper, is unsupported.
minor comments (4)
  1. [Section 4, Eq. before Proposition 1] The polynomials T_k are called Chebyshev polynomials of the second kind, but with T_0=0, T_1=1 they are the shifted Chebyshev polynomials U_{k-1}(λ/2), not the standard U_k(λ). This nomenclature is likely the source of the wrong orthogonality interval and weight used in Section 6.1.
  2. [Section 6.1, proof of Theorem 8] The notation 'L²(-1,1;dρ)' should be 'L²(-1,1;dν)' where dν(λ)=dλ/√(1-λ²), and the statement 'F=0 on (-1,1)' should be stated in the measure sense.
  3. [Remark 5] There is a typo: 'consequntly' should be 'consequently'.
  4. [Section 6.1, Theorem 7 statement] The quantity l^{-1}(z)=Σ|p_k(z)|² appears as a pointwise series; the paper does not discuss convergence of this series for real z, which is relevant because the integral in (26) is over (-1,1).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the paper's main gaps are correctness issues (a false Chebyshev orthogonality identity), not circular reasoning.

full rationale

The paper's central construction (Sections 5–6) does not reduce to its own inputs by definition. The finite-dimensional de Branges space B_T^A is equipped with the norm [F,G] = (C_T f,g), and Eq. (21) derives the equality with ∫ FG dρ via the Fourier transform and finite propagation speed; this is a derived identity, not an assumed one. The semi-infinite space B_∞^A is introduced by analogy, with the scalar product set equal to C[f,g] and then asserted equal to ∫ FG dM using the spectral representation of the response vector r_t = ∫ T_t dM. This latter equality is a substantive theorem from the authors' prior work, and while heavily cited, the prior papers contain proofs; there is no indication that the present argument is a self-citation chain substituting for proof. The proof of Theorem 7 relies on the assertion that the polynomials T_k are orthogonal with respect to dν(λ)=χ_{(-1,1)}(λ)/√(1−λ²)dλ, which is false (for T_1=1 and T_3=λ²−1, ∫ T_1T_3 dν = −π/2 ≠ 0). The same incorrect identity is used in Theorem 8 to convert l² convergence into L²(−1,1;dν) convergence. This is a serious mathematical error that invalidates the proofs as written, but it is not circularity: the false identity is not an input or fitted parameter, and the claims do not become true by definition. Similarly, the statement that the conditions of Theorem 1 are 'trivially checked' is unsupported, but unsupported assertion is a completeness gap, not a circular reduction. No step in the derivation chain is equivalent to its own conclusion by construction.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard de Branges and moment problem theory. The problematic assumptions are the Chebyshev orthogonality identity and the closability and analyticity arguments in Theorem 8, both of which are load-bearing and both of which fail or are incomplete.

assumptions (4)
  • domain assumption T_k(lambda) are orthogonal with respect to dnu(lambda)=chi_(-1,1)(lambda)/sqrt(1-lambda^2) dlambda
    Used in the proof of Theorem 7 (Eq. (28)) and Theorem 8 to identify ||f||^2 with int |F|^2 dnu; false for the T_k defined by T_{t+1}+T_{t-1}-lambda T_t=0, T_1=1.
  • domain assumption The limit in L2(R,M) of polynomials Bf^(n) with f^(n)->0 in l2 lies in the space E of entire functions with reproducing kernel (9)
    Needed for Theorem 8; the proof relies on the failed isometry and on an unproved analyticity and completeness step.
  • standard math Existence and properties of spectral measures drho_{infinity,h} and the Fourier transform with Parseval identity (18)
    Inherited from Atkinson and Schmuedgen and used throughout Sections 4-6.
  • standard math Moment problem determinacy is characterized by lambda_N -> 0 (Theorem 3 of Berg-Chen-Ismail)
    Used in Lemma 1 to conclude the limit point case from the minimal eigenvalue tending to zero.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Inverse problem for semi-infinite Jacobi matrices and associated Hilbert spaces of analytic functions." pith.science (2026). https://pith.science/paper/E5X4LLUP

@misc{pith2026250508338,
  author       = {Pith},
  title        = {Pith review of: Inverse problem for semi-infinite Jacobi matrices and associated Hilbert spaces of analytic functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E5X4LLUP}},
  note         = {Machine review of arXiv:2505.08338}
}
read the original abstract

We consider the dynamic problems for the discrete systems with discrete time associated with finite and semi-infinite Jacobi matrices. The result of the paper is a procedure of association of special Hilbert spaces of functions, namely de Branges space, playing an important role in the inverse spectral theory, with these systems. %Thus the procedure, offered by the authors in the %previous papers now is extended to the case of semi-infinite %Jacobi matrices. We point out the relationships with the classical moment problems theory and compare properties of classical Hankel matrices associated with moment problems with properties of matrices of connecting operators associated with dynamical systems.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 20 canonical work pages

  1. [1]

    Akhiezer

    N.I. Akhiezer. The classical moment problem.Oliver and Boyd, Edinburgh.1965

    work page 1965

  2. [2]

    F. V. Atkinson. Discrete and Continuous Boundary Problems.New York/London, Academic Press.1964

    work page 1964

  3. [3]

    I.Belishev

    M. I.Belishev. Recent progress in the boundary control method.Inverse Problems, 23, no 5, R1–R67, 2007

    work page 2007

  4. [4]

    Boundary control and tomography of Riemannian manifolds (the BC-method)

    M.I.Belishev. Boundary control and tomography of Riemannian manifolds (the BC-method). Uspekhi Matem. Nauk,72, no. 4, 3–66 (in Russian), 2017

    work page 2017

  5. [5]

    Small eigenvalues of large Hankel matrices: The indetermi- nate case.Mathematica Scandinavica,91, 67–81, 2002

    Berg C., Chen Y., Ismail M.E.H. Small eigenvalues of large Hankel matrices: The indetermi- nate case.Mathematica Scandinavica,91, 67–81, 2002

    work page 2002

  6. [6]

    Inverse of infinite hankel moment matricesSymmetry, Integrability and Geometry: Methods and Applications (SIGMA),14, 2018,

    Berg C., Szwarc R. Inverse of infinite hankel moment matricesSymmetry, Integrability and Geometry: Methods and Applications (SIGMA),14, 2018,

    work page 2018

  7. [7]

    Closable Hankel Operators and Moment Problems.Integral Equations and Operator Theory92, no

    Berg C., Szwarc R. Closable Hankel Operators and Moment Problems.Integral Equations and Operator Theory92, no. 1,5, 2020

    work page 2020

  8. [8]

    Hilbert space of entire functions.Prentice-Hall, NJ.1968

    Louis de Branges. Hilbert space of entire functions.Prentice-Hall, NJ.1968

    work page 1968

Show all 21 references
  1. [9]

    H Dym, H. P. McKean. Gaussian processes, function theory, and the inverse spectral problem. Academic Press, New York etc.1976

    1976

  2. [10]

    Mikhaylov, V.S

    A.S. Mikhaylov, V.S. Mikhaylov. Inverse dynamic problems for canonical systems and de Branges spaces.Nanosystems: Physics, Chemistry, Mathematics,9, no. 2, 215-224, 2018

    2018

  3. [11]

    Mikhaylov, V.S

    A.S. Mikhaylov, V.S. Mikhaylov. Boundary Control method and de Branges spaces. Schr¨ odinger operator, Dirac system, discrete Schr¨ odinger operator.Journal of Mathemat- ical Analysis and Applications,460, no. 2, 927-953, 2018

    2018

  4. [12]

    A. S. Mikhaylov, V. S. Mikhaylov. Dynamic inverse problem for Jacobi matrices.Inverse Problems and Imaging,13, no. 3, 431-447, 2019

    2019

  5. [13]

    A. S. Mikhaylov, V. S. Mikhaylov. On application of the Boundary control method to classical moment problems.Journal of Physics: Conference Series,2092(1):012002, 2021

    2021

  6. [14]

    A. S. Mikhaylov, V. S. Mikhaylov. Inverse problem for dynamical system associated with Jacobi matrices and classical moment problems.Journal of Mathematical Analysis and Ap- plications,487, no. 1, 2020

    2020

  7. [15]

    Mikhaylov, V.S

    A.S. Mikhaylov, V.S. Mikhaylov, On a special Hilbert space of functions associated with the multidimensional wave equation in a bounded domain.IEEE Conference Proceedings: Days on Diffraction’2022.101 – 105, 2022

    2022

  8. [16]

    C. Remling . Schr¨ odinger operators and de Branges spaces.J. Funct. Anal.196, no. 2, 323– 394, 2002

    2002

  9. [17]

    R. V. Romanov. Canonical systems and de Branges spaceshttp://arxiv.org/abs/1408.6022

    arXiv

  10. [18]

    Simon, The classical moment problem as a self-adjoint finite difference operator.,Advances in Math.,137, 82-203, 1998

    B. Simon, The classical moment problem as a self-adjoint finite difference operator.,Advances in Math.,137, 82-203, 1998

    1998

  11. [19]

    Schm¨ udgen The moment problem.Springer International Publishing, Cham,2017

    K. Schm¨ udgen The moment problem.Springer International Publishing, Cham,2017

    2017

  12. [20]

    Yafaev Unbounded Hankel Operators and Moment Problems,Integral Equations and Operator Theory,85, no

    D.R. Yafaev Unbounded Hankel Operators and Moment Problems,Integral Equations and Operator Theory,85, no. 2, 289–301, 2016

    2016

  13. [21]

    D.R. Yafaev Correction to: Unbounded Hankel Operators and Moment Problems (Integral Equations and Operator Theory, (2016), 85, 2, (289-300),)Integral Equations and Operator Theory,91, no. 5, 2019. HILBERT SPACES OF ANALYTIC FUNCTIONS 17 St. Petersburg Department of V.A. Steklo...

    2016

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.