Pith. sign in

REVIEW 2 major objections 4 minor 22 references

Complete boundedness of multiple operator integrals

T0 review · 2 major / 4 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 For any finite family of normal operators, a multiple operator integral is bounded on the Haagerup tensor product of compact-operator spaces if and only if its symbol admits a measurable vector-valued factorization, and in that case the…

desk verdict Theorem 8 is the right result and the proof is mostly sound, but Proposition 7—the unproved bridge to Schur multipliers—is load-bearing and needs to be written down before the paper is fully rigorous. read the letter →

arxiv 1908.07879 v1 pith:D7JHJDA7 submitted 2019-08-21 math.FA

Clément Coine This is my paper · ORCID
classification math.FA MSC 46L0747B49
keywords multipleoperatorintegralscompleteboundednessHaageruptensorproductmultilinearSchurmultipliersfactorizationofsymbolsspacesnormaloperatorsperturbationtheory
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

The paper sets out to characterize exactly when a multiple operator integral, the basic multilinear device used in perturbation theory to express differences and derivatives of functions of normal operators, extends from Hilbert-Schmidt inputs to a bounded map on the Haagerup tensor product of compact-operator spaces. It establishes that boundedness is the same as complete boundedness, and that both are equivalent to a concrete factorization of the symbol: the symbol must equal a pointwise inner product of a vector-valued function with a chain of operator-valued functions. The theorem also gives that the norm of the integral, its completely bounded norm, and the norm of the associated continuous multilinear Schur multiplier all coincide with the infimum product of the factor norms. This is the operator-integral analogue of the known factorization characterization for continuous multilinear Schur multipliers.

What carries the argument

The load-bearing object is the Haagerup tensor product $S_\infty(H)\otimes^h\cdots\otimes^h S_\infty(H)$ (with $n-1$ factors), the operator-space tensor product that detects complete boundedness of multilinear maps. The proof works in three steps: first it shows $\Gamma_{A_1,\dots,A_n}(\varphi)$ is a multilinear module map over the commutants of the von Neumann algebras generated by $A_1$ and $A_n$, which forces boundedness to imply complete boundedness; second, using the unitaries $\rho_i$ that identify $L^2(\lambda_{A_i})$ with spectral subspaces of $H$, it identifies the restriction of $\Gamma$ with the continuous multilinear Schur multiplier $\Lambda(\varphi)$; third, complete boundedness of $\Lambda$ is converted, through the complete isometry $CB(E,B(H,K)) = ((K_c)^*\otimes^h E\otimes^h H_c)^*$, into a functional on a Haagerup tensor product of trace-class spaces, whose factorization yields the pointwise inner-product representation of the symbol.

What would settle it

The unproved bridge is Proposition 7, so the shortest check is computational: for $n=2$, take $H=\mathbb{C}^2$, $A_1=\mathrm{diag}(0,1)$, $A_2=\mathrm{diag}(1,0)$, and a symbol such as $\varphi(s,t)=st$; compute $\Lambda(\varphi)$ from the kernel formula in Section 3.2 and $\Gamma_{A_1,A_2}(\varphi)(X)$ from the definition for an explicit $X$. The claim asserts the two sides of equation (8) are equal for every $X$ and every symbol, so a single mismatch disproves the bridge; if the example agrees, repeat the check with $\varphi=\mathbf{1}_{E\times F}$ for a Borel rectangle.

Watch

Extended reading notes

Core claim

The central result, Theorem 8, states that for any $n\ge 2$, any normal operators $A_1,\dots,A_n$ on a separable Hilbert space $H$, and any bounded Borel symbol $\varphi$ on the product of their spectra, the multiple operator integral $\Gamma_{A_1,\dots,A_n}(\varphi)$ extends to a bounded map on the Haagerup tensor product of $n-1$ copies of $S_\infty(H)$ if and only if it extends completely boundedly, if and only if the associated continuous multilinear Schur multiplier $\Lambda(\varphi)$ is completely bounded, if and only if $\varphi$ factors almost everywhere as $\varphi(t_1,\dots,t_n)=\langle a_1(t_1),[a_2(t_2)\cdots a_{n-1}(t_{n-1})](a_n(t_n))\rangle$, where $a_1$ and $a_n$ take values in separable Hilbert spaces and the intermediate $a_i$ are weak-* measurable operator-valued functions on the spectral measures of the corresponding $A_i$. In that case the norm of $\Gamma$, its completely bounded norm, and the norm of $\Lambda$ all equal the infimum of $\|a_1\|_\infty\cdots\|a_n\|_\infty$ over all such factorizations.

Load-bearing premise

The whole chain of equivalences depends on an identification between the operator integral and a certain kernel-smoothing multiplier map, an identification that is asserted without proof in Proposition 7; if that identification fails for some tuple of operators and symbols, the chain breaks.

Editorial extensions

If this is right

  • Boundedness of a multiple operator integral on the Haagerup tensor product automatically implies complete boundedness, with equal norms.
  • The norm of the integral is the infimum over factorizations of the product of the factor norms, so norm estimates become an optimization over vector-valued functions.
  • The same factorization condition characterizes complete boundedness of the associated continuous multilinear Schur multiplier, so the two subjects are identified by the unitary bridge.
  • For perturbation theory, the theorem supplies a concrete criterion for when the multilinear operator integrals representing higher-order differences of operator functions extend to compact operators.

Reading between the lines

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

  • Because the proof uses the Haagerup tensor product and $S_\infty$, it does not settle the analogous question for Schatten classes $S_p$ with $1\le p<\infty$; a natural extension would test whether some analogous factorization characterizes boundedness there.
  • The separability assumptions on $H$ and on the intermediate Hilbert spaces are used to apply structural theorems, but the mechanism itself looks measure-theoretic, so the same statement may hold for non-separable spaces or for arbitrary families of commuting normal operators; this is an extrapolation, not a claim of the paper.
  • The factorization identity isolates a purely spectral condition on $\varphi$, independent of $H$; this suggests that one could check boundedness numerically by searching for vector-valued functions $a_i$ on the spectra, which may be useful for concrete operator-function estimates.
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

2 major / 4 minor

Summary. The manuscript studies multiple operator integrals Γ_{A_1,...,A_n}(φ) acting on Haagerup tensor products of compact operator spaces over a separable Hilbert space. Its main result, Theorem 8, gives four equivalent conditions for such an integral to be bounded: (i) boundedness of the extension to S∞(H) h⊗ ... h⊗ S∞(H); (ii) complete boundedness of this extension; (iii) complete boundedness of the associated continuous multilinear Schur multiplier Λ(φ); and (iv) a pointwise factorization φ(t_1,...,t_n)=⟨a_1(t_1),[a_2(t_2)...a_{n−1}(t_{n−1})](a_n(t_n))⟩ with a_1∈L∞(λ_{A_1};H_1), a_n∈L∞(λ_{A_n};H_{n−1}), and a_i∈L∞_σ(λ_{A_i};B(H_i,H_{i−1})) for separable Hilbert spaces H_i. The proof proceeds through a module-map argument using Juschenko–Todorov–Turowska's Lemma 3.3, a restriction to spectral subspaces that identifies Γ with the Schur multiplier Λ, a duality/factorization argument via L∞_σ spaces, and a finite-rank approximation in the converse direction. The paper also states equality of the bounded, completely bounded, and factorization norms.

Significance. The result is a natural extension of the Juschenko–Todorov–Turowska characterization of continuous multilinear Schur multipliers to general normal operators, and the automatic passage from boundedness to complete boundedness is a useful theorem. The norm formula relating ‖Γ‖, ‖Γ‖_cb, ‖Λ‖_cb, and the infimum over factorizations is a strong and clean statement. The overall strategy is coherent and uses standard operator-space tools; in particular, the finite-rank approximation in (iv)⇒(ii) is explicit and gives a constructive cb bound. The main reservation is that one structural bridge, Proposition 7, is not proved in the manuscript and is essential for (ii)⇔(iii); once that proof is supplied, the result appears credible.

major comments (2)
  1. [Section 3.2, Proposition 7 (equation (8))] Proposition 7 is the only bridge between the multiple operator integral Γ and the multilinear Schur multiplier Λ, and it is used in the proof of (ii)⇒(iii) to restrict Γ to S∞(H_{i+1},H_i) and to identify the restriction with Λ(φ), including the cb-norm inequality ‖Λ(φ)‖_cb ≤ ‖Γ_{A_1,...,A_n}(φ)‖_cb. The proof is omitted with the remark 'The proof is similar and we leave it to the reader.' This is a load-bearing step: without it the equivalence (ii)⇔(iii) in Theorem 8 is not established as written. The author should supply a complete proof, for example by checking (8) for elementary tensors φ=f_1⊗...⊗f_n, where it follows from ρ_i f(A_i)ρ_i^{-1}=M_{f_i}, and then extending by w*-continuity in φ for fixed S2 kernels. If a proof already appears in [4, Proposition 9] for the analogous two-operator case, a precise reference to the exact statement would also be acceptable, but it should be explicit.
  2. [Section 4, proof of (iii)⇒(iv), page 9] The construction of the quotient map q:S1(L2(Ω))→L1(Ω) is invoked with an unresolved cross-reference 'by (??)'. This map is then used to form Q=q_1⊗...⊗q_n, whose kernel N is identified via Proposition 3(iv), and the vanishing of u on N is what produces the factor v. Although the statement that q is a complete metric surjection is plausible and likely standard, the manuscript does not provide the supporting statement. Please replace the placeholder by the correct equation and either prove the complete metric surjectivity or cite a theorem where it appears.
minor comments (4)
  1. [Introduction and Section 3.2, definition of Λ(φ)] The displayed domain of Λ(φ) on page 1 is written as S2(L2(Ω_{n−1}),L2(Ω_n)) × ... × S2(L2(Ω_1),L2(Ω_2)), which is inconsistent with the convention used after (7) and in Proposition 7, where the first argument lies in S2(E2,E1). Please correct this typo.
  2. [Section 4, proof of (iv)⇒(ii)] In the definition of σ_i^N, the displayed formula writes σ_i^N(X_1) but should read σ_i^N(X_i); also σ_1^N should be A_1^N π_N(X_1). This is a typographical error, and the intended argument is clear from the preceding formula.
  3. [Section 4, proof of (iv)⇒(ii)] In the sentence 'the element [a^i_{kl}]_{1≤k,l≤N} ∈ M_N(L∞(λB)) has a norm less than ‖a_i‖∞', the symbol λB should be λA_i.
  4. [Section 4, proof of (iii)⇒(iv)] After obtaining u = v∘Q, the complete boundedness of v follows from the complete metric surjectivity of Q; this step is implicit and should be stated explicitly for the reader.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the new characterization is derived from independent operator-space factorization and external Schur-multiplier results; the only flagged items are an unproved bridge (Proposition 7) and an unresolved reference, which are rigor gaps rather than circular reductions.

full rationale

Walking the derivation chain in Theorem 8: (i) implies (ii) applies [12, Lemma 3.3] after checking the module-map property; that lemma is external to this paper and does not assume Theorem 8's conclusion. (ii) implies (iii) relies on Proposition 7 in Section 3.2, which identifies Gamma restricted to the subspaces H_i with Lambda via equation (8). This proposition is asserted without proof ('The proof is similar and we leave it to the reader'), and it is load-bearing for the equivalence; however it is a concrete intertwining identity, not a restatement of the target theorem, and the paper does not define any input in terms of the output. The identity is not circular; its omission is a completeness or rigor issue. (iii) implies (iv) re-derives the Juschenko-Todorov-Turowska factorization via the complete metric surjection Q, using Proposition 3(iv) and Proposition 4, with a minor unresolved citation '(??)' for the quotient map q:S1(L2(Omega)) to L1(Omega); again this is a missing reference, not a circular step. (iv) implies (ii) is self-contained: it approximates phi by finite-rank symbols and bounds the cb norm using Wittstock's theorem. The prior paper [4] by the author supplies the multiple-operator-integral framework (Theorem 5, w*-continuity, the rho_i unitaries), but [4] does not contain the boundedness and complete-boundedness characterization proved here, and its results are used as external background rather than as the conclusion under proof. No fitted parameter is renamed as a prediction, and no uniqueness theorem from the authors is invoked to force the choice. Therefore the central claim is not circular.

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

The central claim is built on standard operator space theory (Haagerup tensor product factorization, duality, injectivity), the author's earlier multiple operator integral construction from [4], and the external module-map lemma [12, Lemma 3.3]. No parameters are fitted and no entities are invented; the Hilbert spaces appearing in the factorization are existential rather than posited inputs.

assumptions (6)
  • domain assumption Existence and w*-continuity of multiple operator integrals: Γ_{A_1,...,A_n} extends to a w*-continuous isometry from L∞(∏λ_Ai) into B_{n−1}(S2(H)) (Theorem 5, quoted from [4, Theorem 4 and Proposition 5]).
    This is the working definition of multiple operator integrals; the paper takes it from the author's own prior work and does not prove it here.
  • domain assumption Module-map automatic complete boundedness ([12, Lemma 3.3]): any bounded (D,C)-module map from a Haagerup tensor product of copies of S∞(H) into S∞(H) is completely bounded with the same norm.
    External theorem by Juschenko-Todorov-Turowska; it is the engine of (i)⇒(ii).
  • standard math Standard operator space identifications: (Hc)* h⊗ Kc = S1(H,K), Kc h⊗ (Hc)* = S∞(H,K), and CB(E,B(H,K)) = ((Kc)* h⊗ E h⊗ Hc)* (Proposition 4).
    Used to convert complete boundedness of Λ into a linear functional on S1(E1) h⊗ ... h⊗ S1(En).
  • domain assumption Duality L∞_σ(Ω;E*) = B(L1(Ω), E*) (equation (5) from [9]) and the quotient map S1(L2(Ω)) → L1(Ω) given by taking trace-class kernel to its diagonal evaluation.
    Used in (iii)⇒(iv) to identify the abstract factorization of v with pointwise functions ai.
  • standard math Haagerup tensor product properties: factorization theorem for completely bounded maps (Theorem 1), injectivity, and projectivity of the tensor product.
    Standard operator space theory from Pisier [16] and Effros-Ruan [19]; used throughout.
  • domain assumption For every normal operator A on separable H there is a scalar-valued spectral measure λ_A (with same null sets as the spectral measure), and W*(A) has a separating vector (Conway [7, Cor 14.6, Prop 15.3]).
    Provides the measure spaces and the cyclic vectors for commutants needed in (i)⇒(ii).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Complete boundedness of multiple operator integrals." pith.science (2026). https://pith.science/paper/D7JHJDA7

@misc{pith2026190807879,
  author       = {Pith},
  title        = {Pith review of: Complete boundedness of multiple operator integrals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/D7JHJDA7}},
  note         = {Machine review of arXiv:1908.07879}
}
read the original abstract

In this paper, we characterize the multiple operator integrals mappings which are bounded on the Haagerup tensor product of spaces of compact operators. We show that such maps are automatically completely bounded and prove that this is equivalent to a certain factorization property of the symbol associated to the operator integral mapping. This generalizes a result by Juschenko-Todorov-Turowska on the boundedness of continuous multilinear Schur multipliers.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 20 canonical work pages

  1. [12]

    Juschenko, I

    K. Juschenko, I. G. Todorov and L. Turowska, Multidimensional Operator Multipliers, Transactions of the Amer. Math. Soc., Vol. 361 (9), 4683-4720, 2009. https://doi.org/10.1090/s0002-9947-09-04771-0

    work page doi:10.1090/s0002-9947-09-04771-0 2009

  2. [4]

    When do triple operator integrals take value in the trace class?

    C. Coine, C. Le Merdy, F. Sukochev, When do triple operator integrals take value in the trace cla ss?, arXiv:1706.01662

    work page Pith review arXiv

  3. [1]

    Birman, M

    M. Birman, M. Solomyak, Double Stieltjes operator integrals (Russian), Prob. Math. Phys., Izdat. Leningrad Univ. 1 (1966), 33–67. https://doi.org/10.1007/978-1-4684-7595-1_2

    work page doi:10.1007/978-1-4684-7595-1_2 1966

  4. [2]

    Birman, M

    M. Birman, M. Solomyak, Double Stieltjes operator integrals II (Russian), Prob. Math. Phys., Izdat. Leningrad Univ. 2 (1967), 26–60. https://doi.org/10.1007/978-1-4684-7592-0_3

    work page doi:10.1007/978-1-4684-7592-0_3 1967

  5. [3]

    Birman, M

    M. Birman, M. Solomyak, Double Stieltjes operator integrals III (Russian), Prob. Math. Phys., Izdat. Leningrad Univ. 6 (1973), 27–53

    work page 1973

  6. [5]

    Coine, C

    C. Coine, C. Le Merdy, F. Sukochev, A. Skripka, Higher order S2-differentiability and application to Koplienko trace formula , J. Funct. Anal., https://doi.org/10.1016/j.jfa.2018.09.00

    work page doi:10.1016/j.jfa.2018.09.00 2018

  7. [6]

    Perturbation theory and higher order $\mathcal{S}^p$-differentiability of operator functions

    C. Coine, Perturbation theory and higher order Sp-differentiability of operator functions , arXiv:1906.05585

    work page Pith review arXiv 1906

  8. [7]

    Conway, A Course in Operator Theory , Graduate Studies in Mathematics, Vol

    J. Conway, A Course in Operator Theory , Graduate Studies in Mathematics, Vol. 21. American Math- ematical Society, 2000

    work page 2000

Show all 22 references
  1. [8]

    Diestel, J

    J. Diestel, J. J. Uhl, Vector Measures, Mathematical Surveys 15, American Mathematical Society, 1979

    1979

  2. [9]

    Dunford, B

    N. Dunford, B. Pettis, Linear operator on summable functions , Trans. Amer. Math. Soc. 47 (1940), 323-392. https://doi.org/10.1090/s0002-9947-1940-0002020-4 COMPLETE BOUNDEDNESS OF MULTIPLE OPERATOR INTEGRALS 13

    1940 doi

  3. [10]

    E. G. Effros, Zh.-J. Ruan, Multivariable multipliers for groups and their operator al gebras. Operator theory: operator algebras and applications, Part 1 (Durham, NH, 1988), 197–218, Proc. Sympos. Pure Math., 51, Part 1, Amer. Math. Soc., Providence, RI, 1990

    1988

  4. [11]

    Caspers, G

    M. Caspers, G. Wildschut, On the complete bounds of Lp-Schur multipliers , Arch. Math. (2019) 113:

    2019

  5. [13]

    Le Merdy, A

    C. Le Merdy, A. Skripka, Higher order differentiability of operator functions in Sch atten norms , J. Inst. Math. Jussieu, https://doi.org/10.1017/s1474748019000033

    doi

  6. [14]

    Pavlov, Multidimensional operator integrals (Russian), Problems of Math

    B. Pavlov, Multidimensional operator integrals (Russian), Problems of Math. Anal., No. 2: Linear Op- erators and Operator Equations (Russian), (1969) 99–122

    1969

  7. [15]

    V. V. Peller, Multiple operator integrals and higher operator derivativ es, J. Funct. Anal. 233 (2006), no. 2, 515-544. https://doi.org/10.1016/j.jfa.2005.09.003

    2006 doi

  8. [16]

    Pisier, Introduction to operator space theory , London Mathematical Society, Lecture note Series 294, 2003

    G. Pisier, Introduction to operator space theory , London Mathematical Society, Lecture note Series 294, 2003

    2003

  9. [17]

    Pisier , Similarity problems and completely bounded maps , Lecture Notes in Mathematics, 1618

    G. Pisier , Similarity problems and completely bounded maps , Lecture Notes in Mathematics, 1618. Springer-Verlag, Berlin, 1996

    1996

  10. [18]

    Potapov, A

    D. Potapov, A. Skripka, F. Sukochev, Spectral shift function of higher order , Invent. Math. 193 (2013), no. 3, 501–538. https://doi.org/10.1007/s00222-012-0431-2

    2013 doi

  11. [19]

    E. G. Effros and Z. Ruan, Operator spaces, London Mathematical Society Monographs New Series 23, 2000

    2000

  12. [20]

    Spronk, Measurable Schur multipliers and completely bounded multi pliers of the Fourier algebras , Proc

    N. Spronk, Measurable Schur multipliers and completely bounded multi pliers of the Fourier algebras , Proc. London Math. Soc. (3) 89(2004), 161–192. https://doi.org/10.1112/s0024611504014650

    2004 doi

  13. [21]

    Stenkin, Multiple operator integrals , Izv

    V.V. Stenkin, Multiple operator integrals , Izv. Vysh. Uchebn. Zaved. Matematika no. 4 (1977), 102115 (Russian). English transl.: Soviet Math. (Iz. VUZ) 21:4 (1977), 8899. E-mail address : clement.coine1@gmail.com School of Mathematics and Statistics, Central South Univer sity...

    1977

  14. [189]

    https://doi.org/10.1007/s00013-019-01316-7

    doi

Pith tools

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