Pith. sign in

REVIEW 2 major objections 3 minor 12 references

Complete functional calculus bounds for $\rho$-contractions

T0 review · 2 major / 3 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read The paper establishes that for any ρ-contraction T on a Hilbert space, a contractive matrix-valued analytic function F satisfies an explicit sharp bound determined solely by the matrix F(0).

desk verdict The matrix-valued bound is a real advance, but Proposition 3.5 has a wrong factorization in the central positivity argument—fixable, but the proof as printed is not correct. read the letter →

arxiv 2607.13794 v1 pith:NWVFJN4O submitted 2026-07-15 math.FA

Michael Hartz , Jens de Vries This is my paper · ORCID
classification math.FA MSC 47A6047A2047A2547A3047A63
keywords ρ-contractionfunctionalcalculuscompleteboundPotapov–Möbiustransformsmallestsingularvalueunitarydilationspectralsetmatrix-valuedanalyticfunction
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 proves sharp complete functional calculus bounds for ρ-contractions: for every contractive matrix-valued analytic function F on the unit disc, the norm of F(T) is bounded by an explicit decreasing function of the data at zero. When ρ is at most 1, the bound depends on the operator norm of F(0); when ρ exceeds 1, it depends on the smallest singular value of F(0). The result is the matrix-valued refinement of the scalar bounds of Drury and of Schwenninger and de Vries, and it sharpens the classical complete ρ-spectral set bound of Okubo and Ando. The authors also show that the main estimate is optimal, both at the level of the norm inequality and, under natural assumptions, at the level of the underlying operator inequality.

What carries the argument

The argument combines Potapov–Möbius transforms M_A on the unit ball of M_n(C) with the new operator class C_ρ^{(n)}: ρ-contractions that remain of class C_ρ after multiplication by arbitrary pairs of contractive scalar matrices. Lemma 2.1 shows that composing a ρ-contraction with a matrix-valued function vanishing at 0 produces an operator in C_ρ^{(n)}. Lemma 3.3 reduces the operator inequality M_A(T)*M_A(T) ≤ B to the positivity of a 2×2 matrix W(t) on the spectrum of A*A, and Proposition 3.5 exhibits the explicit b_{ρ,K} for which this positivity holds, choosing the auxiliary parameter x(t) to cancel off-diagonal terms on the affine branch and setting x(t) = −1 on the non-affine branch. T

What would settle it

A direct computer-algebra check of the 2×2 matrix W(t) from Lemma 3.3 with b = b_{ρ,K}, δ = ρK, and the proposed x(t) would settle the key step: for example, take ρ = 2, K = 1.5, and t = 0.9 (which lies on the non-affine branch) and verify whether det(W(t)) is non-negative; a negative determinant would contradict Proposition 3.5 and hence Theorem 1.2. Verifying identity (7) for generic ρ, K, and t on the affine branch would test the affine part.

Watch

Extended reading notes

Core claim

Theorems 1.1 and 1.2 give the sharp complete bound: for T of class C_ρ and ||F||_∞ ≤ 1, ||F(T)|| ≤ c_ρ(||F(0)||²) when ρ ≤ 1 and ||F(T)|| ≤ c_ρ(m(F(0))²) when ρ > 1, where c_ρ(t) = (ρ/2)(1−t) + sqrt(ρ²(1−t)²/4 + t) and m is the smallest singular value. At the operator level, F(T)*F(T) is bounded above by an explicit function b_{ρ,K}(A*A), defined piecewise with an affine branch and, for ρ > 1, a non-affine branch. The bound is attained for every admissible matrix F(0) by an explicit example, and Theorem 5.3 shows that among bounds of the form h(F(0)*F(0)) with the natural normalization, the function b_{ρ,K} is optimal.

Load-bearing premise

The claimed operator bound follows only if the algebraic factorisations in Proposition 3.5 — the identity (7) on the affine branch and the w(s)w(s)* splitting on the non-affine branch — are correct; these steps are asserted with computations not fully expanded in the text, and the proof of Theorem 1.2 depends on them.

Editorial extensions

If this is right

  • If Theorem 1.1 is correct, the Okubo–Ando complete ρ-spectral set bound follows as a corollary, and combining it with Paulsen's similarity theorem yields a new proof of the Okubo–Ando similarity result.
  • For n = 1, the bound reduces to the scalar refined estimate ||f(T)|| ≤ c_ρ(|f(0)|²), recovering the earlier results of Drury and of Schwenninger and de Vries.
  • For ρ = 2, the class C_2 consists of operators with numerical radius at most 1, so the theorem provides a matrix-valued von Neumann-type inequality with constant c_2(m(F(0))²), which can be strictly smaller than the classical constant 2.
  • Proposition 5.1 shows that the constants in Theorem 1.1 cannot be improved for any fixed matrix A = F(0), settling sharpness of the norm-level bound.
  • Theorem 5.3 shows that the operator-level bound b_{ρ,K}(A*A) is optimal among all bounds of the form h(F(0)*F(0)) that are continuous, normalized by h(t₀) = K², and valid for all matrix-valued functions with the prescribed value at zero.

Reading between the lines

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

  • The switch from the operator norm of F(0) for ρ ≤ 1 to the smallest singular value for ρ > 1 suggests a geometric interpretation: for ρ > 1, the functional calculus is controlled by the lowest-rank direction of the initial value, which may have consequences for numerical-radius inequalities for matrix-valued functions.
  • The piecewise definition of b_{ρ,K} indicates a genuine dichotomy that might reflect two distinct regimes in the action of Potapov–Möbius transforms; a natural testable extension would be to determine whether an optimal bound with a smoother function could exist if one drops the assumption that the bound depends only on F(0)*F(0).
  • The matrix-positivity criterion of Lemma 3.3 is phrased spectrally and might adapt to commuting tuples of ρ-contractions or to functions of several variables, where sharp complete bounds are still largely open.
  • For ρ < 1, the bound depends on the largest singular value and the function c_ρ is increasing, so the worst case occurs when F(0) = 0; the symmetry between ρ and 1/ρ suggests a duality relation that could be made explicit by comparing the two branches of b_{ρ,K}.
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

2 major / 3 minor

Summary. The paper proves complete (matrix-valued) functional calculus bounds for ρ-contractions. For F in M_n(O(D)) with ||F||_∞ ≤ 1, Theorem 1.1 bounds ||F(T)|| by c_ρ(||F(0)||^2) for ρ ≤ 1 and by c_ρ(m(F(0))^2) for ρ > 1, refining the Okubo–Ando complete spectral set bound. The more precise Theorem 1.2 gives an operator upper bound F(T)^*F(T) ≤ b_{ρ,K}(A^*A), where b_{ρ,K} is an explicit piecewise function. The proof introduces a class C_ρ^(n) of matrix-intertwined ρ-contractions and uses Potapov–Möbius transforms; the key algebraic step is Proposition 3.5, which verifies a positivity criterion for W(t). Section 5 addresses sharpness, including a lower bound for any admissible h and an optimality theorem under natural assumptions.

Significance. If the main theorems hold, this is a substantial refinement of the Okubo–Ando bound in the complete setting, and the first sharp complete functional calculus bound for ρ-contractions with a data-dependent constant. The proof strategy via Potapov–Möbius transforms and the class C_ρ^(n) is natural and likely correct after the algebraic correction noted below. The paper is transparent about AI-assisted computations, which is commendable but also means that all such identities require independent verification. The sharpness discussion in Section 5 is valuable, though it rests on a nontrivial computation that should be made fully checkable.

major comments (2)
  1. [Proposition 3.5, non-affine branch] The displayed factorization of W(t) is algebraically false. Direct expansion from Lemma 3.3 with δ=ρK and x(t)=−1 gives W(t) = 1/(d(1−s)) w(s)w(s)^*, where d=2(K−1)−(ρK−1)(1−s). The printed formula instead has the prefactor ((1−s†)+(s−s†))/((ρK−1)(1−s)). Since d=(ρK−1)((1−s†)+(s−s†)), the printed prefactor is off by a factor ((1−s†)+(s−s†))². For ρ=2, K=1.5, t=0.64, the left side is [[1/12,1/6],[1/6,1/3]] while the printed right side is [[3/400,3/200],[3/200,3/100]]. The corrected prefactor is positive on the stated branch, so the conclusion W(t)≥0 remains true, but the proof as written contains an incorrect equality at a load-bearing step. The authors should correct the identity and re-check that no later argument depends on the erroneous prefactor.
  2. [Theorem 5.3, proof] The step from the Schur complement to inequality (10), and the subsequent claim that the last summand of (10) can be made to vanish by choosing λ∈[0,∞), is a substantial algebraic simplification that is not shown. Since this is the basis for the sharpness claim, the paper should provide the full expansion or a verifiable symbolic computation. The present text is not independently checkable from the material given, especially in light of the AI-assisted provenance of part of this construction.
minor comments (3)
  1. [Theorem 5.3, proof] The displayed matrix A is malformed. From the subsequent compression of h(A^*A), A should be a 2×2 diagonal matrix with singular values √t and √t0 (e.g., diag(√t,√t0) or diag(√t0,√t)), not the matrix shown in the text.
  2. [Lemma 2.3] The sentence 'the spectral radius of an operator of class C_ρ is at most ρ/(2−ρ)' is imprecise as stated; the cited bound applies for ρ<1, whereas for ρ≥1 the spectral radius is at most 1 by the unitary dilation. Consider clarifying the distinction.
  3. [Proposition 5.2] The notation h(F(0)^*F(0)) for a matrix-valued argument should be explicitly defined as the continuous functional calculus applied to the normal matrix F(0)^*F(0), since h is a scalar function.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the matrix functional calculus bounds are proved by a self-contained construction, with self-citations only as context.

full rationale

I followed the derivation chain from Theorems 1.1 and 1.2 back through Proposition 3.5 and Lemma 3.3. Theorem 1.2 is obtained by decomposing F as M_A ∘ G and applying Proposition 3.5; Proposition 3.5 verifies the sufficient positivity condition of Lemma 3.3 for an explicitly defined function b_{ρ,K}. The parameters K = c_ρ(m(A)^2) or c_ρ(‖A‖^2) are not fitted to the conclusion; they are determined by the quadratic equation defining c_ρ and are shown sharp by explicit examples in Section 5. The prior results [11], [7], and [2] are used for context, for the scalar prototype, or for a special case of weighted shifts; the matrix bound itself is proved directly in this paper. The disputed algebraic factorization in Proposition 3.5 would be a correctness concern, not a circularity, since the claimed inequality does not reduce to its input by construction. No parameter is fitted to the predicted quantity, and no uniqueness theorem from the authors is invoked to force the choice. The self-citations are not load-bearing in the proof of the main estimate.

Assumptions & free parameters 0 free parameters · 6 assumptions · 1 invented entities

No numbers are fitted to data. The paper constructs explicit functions c_rho and b_{rho,K} with closed forms; the parameter K is determined by the singular values of A, and lambda in Theorem 5.3 is a proof device chosen to force a term to vanish. The only newly introduced object is the class C_rho^(n), which is defined and characterized within the paper.

assumptions (6)
  • standard math A contraction on a Hilbert space has a unitary dilation (Sz.-Nagy).
    Used in Lemma 2.1 to dilate G(U) to a unitary 1-dilation; standard theorem, see [12].
  • standard math Operators in C_rho have spectral radius at most rho/(2-rho).
    Used in Lemma 2.3 for rho<1 to show the spectral radius of YT is at most 1; cited to [1, Lemma 5] and [6, Theorem 1].
  • standard math Pluriharmonic minimum principle on operator balls / reduction to unitaries.
    Used in Lemma 2.3 to reduce the positivity of a pluriharmonic map to unitary X; standard complex analysis.
  • standard math Potapov-Mobius transforms map the unit ball into itself and factor any contractive F as M_A composed with G with G(0)=0.
    Central decomposition in Section 2; follows from known theory (see [7], [8, Theorem 8.19]).
  • standard math Paulsen similarity theorem: a complete rho-spectral set implies similarity with ||S|| ||S^{-1}|| <= rho.
    Used in the introduction to note Theorem 1.1 recovers Okubo-Ando's similarity result; cited [10].
  • standard math Sylvester's criterion for positive semidefinite matrices.
    Used in Lemma 5.4 to verify the C_rho characterization (1) for a 3x3 weighted shift.
invented entities (1)
  • C_rho^(n) — class of n-matrix-intertwined rho-contractions independent evidence
    purpose: A subclass of C_rho stable under multiplication by scalar matrices; the proof analyzes Potapov-Mobius transforms on this class.
    Defined in Section 2, characterized in Remark 2.2 (scalar multiples of identity belong iff contractive), and instantiated by Lemma 2.1. It is a working tool with proven properties, not a hat-pulled entity.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Complete functional calculus bounds for $\rho$-contractions." pith.science (2026). https://pith.science/paper/NWVFJN4O

@misc{pith2026260713794,
  author       = {Pith},
  title        = {Pith review of: Complete functional calculus bounds for $\rho$-contractions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NWVFJN4O}},
  note         = {Machine review of arXiv:2607.13794}
}
abstract

Let $T$ be a $\rho$-contraction on a Hilbert space. We establish sharp bounds for the functional calculus of $T$ for matrix-valued analytic functions on the unit disc. These are complete versions of bounds established by Drury and by Schwenninger and the second author.

Figures

Figures reproduced from arXiv: 2607.13794 by the authors.

Figure 1
Figure 1. Graphs of cρ for ρ = 1 2 , 2, 3. We remark that in the case ρ ≥ 1, Theorem 1.1 gives (yet another) proof of the bound (4) of Okubo and Ando, and hence of their similarity result, thanks to Paulsen’s similarity theorem. Theorem 1.1 is in fact a consequence of the following more precise result, which provides an operator upper bound on F(T) ∗F(T) that only depends on F(0). Theorem 1.2. Let ρ > 0 and let T ∈ Cρ. Let F … view at source ↗
Figure 2
Figure 2. The graphs of bρ,K for 0 < ρ < 1 and K = cρ(∥A∥ 2 ) in the cases K = ρ (i.e. ∥A∥ = 0) and K > ρ (i.e. ∥A∥ > 0). Now let ρ > 1. Then for any fixed t, the function K 7→ bρ,K(t) is increasing on [1, ρ]. The loosest upper bound occurs at K = ρ. If K = cρ(m(A) 2 ) as in Proposition 3.5, then K = ρ is equivalent to m(A) = 0. Conversely, as m(A) increases, the value of K decreases, resulting in a tighter upper bound bρ,K(t… view at source ↗
Figure 3
Figure 3. The graphs of bρ,K for ρ > 1 and K = cρ(m(A) 2 ) in the cases K = ρ (i.e. m(A) = 0) and K < ρ (i.e. m(A) > 0). Remark 3.6. We briefly comment on the choices of parameters in the proof of Proposition 3.5 in the case ρ > 1. Essentially, we minimized b(t) subject to the constraints given by Lemma 3.3 (i.e. there exists a δ ≥ 0 and a function x: σ(A∗A) → [−1, 1] such that W(t) ≥ 0 for all t ∈ σ(A∗A)). In particular, we … view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 1 linked inside Pith

  1. [11]

    Schwenninger and Jens de Vries,A sharp bound for the functional calculus ofρ-contractions, Proc

    Felix L. Schwenninger and Jens de Vries,A sharp bound for the functional calculus ofρ-contractions, Proc. Amer. Math. Soc.153(2025), no. 11, 4759–4768

    2025

  2. [1]

    C. A. Berger and J. G. Stampfli,Norm relations and skew dilations, Acta Sci. Math. (Szeged)28(1967), 191–195

    1967

  3. [2]

    Laurent Carrot,Computation of theρ-numerical radius for truncated shifts, Acta Sci. Math. (Szeged)69(2003), no. 1-2, 311–322

    2003

  4. [3]

    Operator Theory90(2023), no

    Raphaël Clouâtre, Maëva Ostermann, and Thomas Ransford,An abstract approach to the Crouzeix conjecture, J. Operator Theory90(2023), no. 1, 209–221

    2023

  5. [4]

    S. W. Drury,Symbolic calculus of operators with unit numerical radius, Linear Algebra Appl.428(2008), no. 8-9, 2061–2069

    2008

  6. [5]

    Eckstein and A

    G. Eckstein and A. Rácz,Weighted shifts of classCρ, Acta Sci. Math. (Szeged)35(1973), 187–194

    1973

  7. [6]

    Japan Acad.44(1968), 16–20

    Takayuki Furuta,Relations between unitaryρ-dilatations and two norms, Proc. Japan Acad.44(1968), 16–20

    1968

  8. [7]

    McCarthy,From Clouâtre-Ostermann-Ransford to Okubo-Ando, (2026), arXiv:2606.02922

    Michael Hartz and John E. McCarthy,From Clouâtre-Ostermann-Ransford to Okubo-Ando, (2026), arXiv:2606.02922

    arXiv 2026

Show all 12 references
  1. [8]

    Isidro and László L

    José-M. Isidro and László L. Stachó,Holomorphic automorphism groups in Banach spaces: an elementary intro- duction, North-Holland Mathematics Studies, vol. 105, North-Holland Publishing Co., Amsterdam, 1985, Notas de Matemática, 97. [Mathematical Notes]

    1985

  2. [9]

    4, 385–394

    Kazuyoshi Okubo and Tsuyoshi Ando,Constants related to operators of classCρ, Manuscripta Math.16(1975), no. 4, 385–394

    1975

  3. [10]

    Paulsen,Completely bounded homomorphisms of operator algebras, Proc

    Vern I. Paulsen,Completely bounded homomorphisms of operator algebras, Proc. Amer. Math. Soc.92(1984), no. 2, 225–228

    1984

  4. [12]

    F achrichtung Mathematik, Universität des Saarlandes, 66123 Saarbrücken, Germany Email address:hartz@math.uni-sb.de Email address:jens_de_vries@math.uni-sb.de

    Béla Sz.-Nagy, Ciprian Foias, Hari Bercovici, and László Kérchy,Harmonic analysis of operators on Hilbert space, second ed., Universitext, Springer, New York, 2010. F achrichtung Mathematik, Universität des Saarlandes, 66123 Saarbrücken, Germany Email address:hartz@math.uni-sb...

    2010

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.