Pith. sign in

REVIEW 3 major objections 5 minor 35 references

Commutator estimates for functions of noncommuting self-adjoint operators

T0 review · 3 major / 5 minor · reviewed 2026-08-27 · deepseek-v4-flash

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

Pith's one-line read This paper proves a sharp threshold: the functional calculus for pairs of noncommuting self-adjoint operators satisfies commutator Lipschitz and multiplicativity estimates in the Schatten class $S_p$ exactly for $1\le p\le 2$, and these…

desk verdict Solid negative half, plausible but under-proved positive half with misstated inequalities; deserves a specialist referee. read the letter →

arxiv 2608.16731 v1 pith:ZNHBLZPW submitted 2026-08-17 math.FA math.CAmath.CVmath.SP

Vladimir Peller This is my paper · ORCID
classification math.FAmath.CAmath.CVmath.SP MSC 47A6047B1547B1047B47
keywords functionalcalculusself-adjointoperatorscommutatorsSchatten–vonNeumannclassesdoubleoperatorintegralstripleHaageruptensorproductsBesovclass
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 establish an exact threshold for how well the functional calculus $\varphi\mapsto\varphi(A,B)$ of two noncommuting self-adjoint operators can be controlled by commutators. It proves that for $1\le p\le 2$ the commutator $[\varphi(A,B),Q]$ lies in the Schatten--von Neumann class $S_p$ and its norm is bounded by a constant times the Besov norm of $\varphi$ and the $S_p$ norms of $[A,Q]$ and $[B,Q]$; it also proves that when $[A,B]\in S_p$ the calculus is multiplicative modulo $S_p$, meaning $[\varphi(A,B),\psi(A,B)]$ is controlled by $[A,B]$ in $S_p$. For $p>2$, it proves that no such commutator Lipschitz estimate can exist, even for finite-rank self-adjoint operators, and that the same failure occurs in the operator norm. The divide matters because it shows the range $p\le 2$ is the maximal regime in which the calculus inherits Lipschitz-type stability from the input operators.

What carries the argument

The machinery consists of representing the commutator $[\varphi(A,B),Q]$ as the sum of two triple operator integrals, one with symbol the divided difference $D^{[1]}\varphi(x_1,x_2,y)$ and one with $D^{[2]}\varphi(x,y_1,y_2)$. For $\varphi$ in the Besov class $B^1_{\infty,1}(\mathbb{R}^2)$, both divided differences lie in the Haagerup-like tensor products $L^\infty\otimes^h L^\infty\otimes^h L^\infty$ of the first and second kind, and inequalities (2.3) and (2.5) bound the resulting triple operator integrals in $S_p$ for $1\le p\le 2$. The negative direction uses Lemma 4.3, a block matrix embedding that converts a hypothetical commutator Lipschitz estimate into the one-variable perturbation estimate $\|\varphi(A_1,B)-\varphi(A_2,B)\|_{S_p}\le K\|\varphi\|_{B^1_{\infty,1}}\|A_1-A_2\|_{S_p}$, which is known from [ANP] to fail for $p>2$.

What would settle it

Fix $1<p<2$, take finite-rank self-adjoint $A,B$ and a bounded $Q$ with $[A,Q],[B,Q]\in S_p$, and compute the ratio $\|[\varphi(A,B),Q]\|_{S_p}/(\|\varphi\|_{B^1_{\infty,1}}(\|[A,Q]\|_{S_p}+\|[B,Q]\|_{S_p}))$ over a family of growing rank. An unbounded sequence would refute Theorem 3.1; equivalently, a direct verification that inequality (2.3) or (2.5) fails for some $1<p<2$ would invalidate the proof of the positive claim.

Watch

Extended reading notes

Core claim

The central discovery is a dichotomy ruled by $p$. When $1\le p\le 2$ and $A,B,Q$ are such that $[A,Q],[B,Q]\in S_p$, the paper proves for every $\varphi\in B^1_{\infty,1}(\mathbb{R}^2)$ the commutator bound $\|[\varphi(A,B),Q]\|_{S_p}\le C\|\varphi\|_{B^1_{\infty,1}}(\|[A,Q]\|_{S_p}+\|[B,Q]\|_{S_p})$, and when $[A,B]\in S_p$ it proves $\|[\varphi(A,B),\psi(A,B)]\|_{S_p}\le C\|\varphi\|_{B^1_{\infty,1}}\|\psi\|_{B^1_{\infty,1}}\|[A,B]\|_{S_p}$. For $p>2$, it shows no constant $K$ can make the first inequality hold uniformly over finite-rank self-adjoint $A,B$ and bounded $R$, and Theorem 4.2 shows the same in the operator norm. The negative results are obtained by a block-matrix construction that would turn any such commutator bound into a one-variable perturbation estimate, which is already known to fail for $p>2$.

Load-bearing premise

The proof of the positive results assumes without reproof the Schatten--von Neumann bounds (2.3) and (2.5) for triple operator integrals of the first and second gender in the range $1\le p\le 2$; if either bound fails at some $p$, Theorem 3.1 collapses, and the negative results also depend on the previously established failure of the one-variable perturbation estimate for $p>2$.

Editorial extensions

If this is right

  • For every $\varphi\in B^1_{\infty,1}(\mathbb{R}^2)$ and every $Q$ with $[A,Q],[B,Q]\in S_p$, the commutator $[\varphi(A,B),Q]$ is itself in $S_p$ and its norm depends only on the Besov norm of $\varphi$ and the two input commutator norms.
  • When $A$ and $B$ almost commute, i.e. $[A,B]\in S_1$, the commutator $[\varphi(A,B),\psi(A,B)]$ is trace class and its norm satisfies the multiplicative bound with $\|[A,B]\|_{S_1}$, extending the earlier $p=1$ result.
  • For $p>2$, no commutator Lipschitz bound holds in $S_p$ or in the operator norm, so the threshold $p=2$ is sharp and the positive results cannot be pushed past it.
  • The block-matrix reduction in Lemma 4.3 implies that any future commutator Lipschitz estimate for pairs at some $p$ would automatically yield the corresponding one-variable perturbation estimate, linking the two problems.

Reading between the lines

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

  • A natural testable extension is whether the same $p\le 2$ dichotomy holds for functional calculi of three or more noncommuting operators; the two-variable failure for $p>2$ suggests the multivariate versions will also fail beyond $p=2$.
  • The negative results imply that numerical schemes approximating $\varphi(A,B)$ for noncommuting $A,B$ by commutator expansions should only trust $S_p$ error bounds for $p\le 2$; for $p>2$, small input commutators do not guarantee small output commutators.
  • Since the one-variable perturbation estimate for $p>2$ is the engine of the negative results, the dichotomy can be read as: the two-variable commutator problem is exactly as hard as the one-variable perturbation problem, so the threshold is inherited rather than new.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

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 double-operator-integral functional calculus φ ↦ φ(A,B) for self-adjoint operators A,B whose commutator lies in a Schatten–von Neumann class S_p. The main claim is a dichotomy: for 1 ≤ p ≤ 2, the calculus on the homogeneous Besov class B^1_{∞,1}(R^2) satisfies a commutator Lipschitz estimate in S_p and is multiplicative modulo S_p, while for p > 2 no such estimate holds in S_p or in the operator norm, even for finite-rank self-adjoint A and B. The positive results are stated as Theorems 3.1 and 3.2, whose proofs are delegated to the author's prior work [AP2] together with inequalities (2.3) and (2.5) for triple operator integrals. The negative results are derived in Section 4 via a block-matrix reduction (Lemma 4.3) that converts failure of the one-variable perturbation estimate from [ANP, §8] into failure of the two-variable commutator estimate.

Significance. If the results are correct, the paper gives a sharp border p = 2 for commutator Lipschitz estimates of functions of noncommuting self-adjoint operators, extending previous results from p = 1 to the whole range 1 ≤ p ≤ 2 and providing clean counterexamples for p > 2. The negative half is supported by a simple and convincing block-matrix argument that reduces to a known one-variable result, and this part appears sound modulo a statement typo. The positive half, however, is not proved in the manuscript: it rests entirely on inequalities that are stated inconsistently and whose applicability to the specific terms in Theorems 3.1 and 3.2 is not demonstrated. Because the central dichotomy depends on both halves, the positive half needs a complete and corrected proof before the paper can be accepted.

major comments (3)
  1. [§2, §3, Eqs (2.3), (2.5), Theorem 3.1] The proof of Theorem 3.1 is not actually supplied; the sentence 'The proof of Theorem 3.1 can be obtained by following the line of the proof of Theorem 4.1 of [AP2] and applying inequalities (2.3) and (2.5)' delegates the central positive claim to inequalities that are stated inconsistently in §2. In the definition of first-gender integrals, T is bounded and R∈S_p, but (2.3) bounds by ∥T∥_{S_p}∥R∥; in the second-gender definition T is bounded and R∈S_p, and (2.5) bounds by ∥T∥∥R∥_{S_p}. Moreover, the two terms displayed in Theorem 3.1 need opposite placements of the S_p factor: the first term has [B,Q]∈S_p in the final position, while the second has [A,Q]∈S_p in the middle position. The manuscript never states which of (2.3) or (2.5) is meant to cover which term, and the configuration with the S_p operator in the middle and the identity in the final position is not covered by either displayed definition. Since this is the entire support for the p≤2 positive half, the authors must state the correct inequalities, either prove them or give precise published references with matching hypotheses, and spell out how each term in Theorems 3.1 and 3.2 is bounded.
  2. [Lemma 4.3, Eq. (4.2)] The conclusion of Lemma 4.3 is misprinted. The proof computes AR−RA with upper-right entry A1−A2 and BR−RB with upper-right entry B1−B2, so the correct conclusion is ∥φ(A1,B1)−φ(A2,B2)∥_{S_p} ≤ K∥φ∥_{B^1_{∞,1}} max{∥A1−A2∥_{S_p}, ∥B1−B2∥_{S_p}}. As printed, with ∥A1−B1∥ and ∥A2−B2∥, the lemma is false in general (for example, take A1=A2 and B1≠B2, then the right-hand side is nonzero while the left-hand side is not forced to vanish). The statement must be corrected, and Theorems 4.1 and 4.2 should be read as relying on the corrected version.
  3. [Theorems 3.1 and 3.2] The hypotheses of Theorems 3.1 and 3.2 say only that A and B are self-adjoint operators, while the cited results from [ANP] and [AP2] concern bounded self-adjoint operators. The condition [A,Q]∈S_p does not imply that A is bounded (for instance, Q can be a finite-rank projection onto vectors in the domain of an unbounded A). Since the proof is delegated to the bounded-operator results in [AP2], the theorems should either assume A and B are bounded or provide an additional approximation argument showing that the unbounded case follows from the bounded case.
minor comments (5)
  1. [Abstract] The abstract contains the typo 'Schaten–von Neumann'; it should be 'Schatten–von Neumann'.
  2. [Eq. (2.3)] In inequality (2.3), the symbol Ψ should be Φ; the displayed norm should refer to the symbol Φ of the triple operator integral.
  3. [Section 3 heading] The heading reads 'Commutator Lipschitz estimates in S_p in the case p∈[1.2]'; the intended interval is p∈[1,2].
  4. [Definition of D[2]φ] In the definition of D[2]φ, the second line should read (∂φ/∂y)(x,y1) for y1=y2, not (∂φ/∂y)(x1,y).
  5. [References] In reference [Pe1], the journal name contains a typo: 'Funk. anal. i ego pril.' should likely be 'Funkts. anal. i ego pril.'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: load-bearing estimates come from prior independent work; the printed (2.3) has a typo, which is a rigor issue rather than a circular reduction.

full rationale

The paper's central claims rest on two kinds of inputs: the triple-operator-integral bounds (2.3)/(2.5) for the positive p<=2 half, and the no-perturbation-Lipschitz result from §8 of [ANP] for the negative p>2 half. Both are quoted from the author's own prior papers ([ANP], [AP2], [AP4]). This is heavy self-citation, but it is not circular: those prior results are parameter-free published theorems with stated hypotheses that do not include the target commutator-Lipschitz estimate or the target multiplicative-modulo-S_p estimate, and the paper does not fit any parameter to data and then rename it a prediction. Theorem 3.1 is a genuine application of the triple-integral estimates to a new commutator identity; Lemma 4.3 converts a hypothetical commutator bound into the perturbation bound already ruled out in [ANP], which is a valid contrapositive rather than an assumption of the conclusion. The manuscript does contain a technical gap in the statement of (2.3): the first-gender definition assumes T bounded and R in S_p, while the displayed inequality is printed with ||T||_{S_p}||R||; and the proof of Theorem 3.1 requires checking that the two asymmetric placements (T=I,R=[B,Q] and T=[A,Q],R=I) are both covered by (2.3)/(2.5), which the text does not spell out. There is also a typo in (4.2), which the proof shows should involve ||A1-A2|| and ||B1-B2||. These are rigor and correctness concerns, not circular reductions. Accordingly the circularity score is 0.

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

The paper introduces no free parameters and no new entities. It imports all technical machinery from the author's earlier program, including double and triple operator integrals, Haagerup-like tensor products, and Besov space embedding. The axioms listed are the load-bearing background assumptions that the central claims depend on, all drawn from prior published work.

assumptions (4)
  • standard math The spectral theorem represents every self-adjoint operator by a spectral measure, allowing the definition of φ(A,B) as a double operator integral.
    This is the foundation of the entire calculus; it is used in the definition of φ(A,B) in Section 1 without proof.
  • domain assumption For φ in the homogeneous Besov class B^1_{∞,1}(R^2), the divided differences D^{[1]}φ and D^{[2]}φ belong to Haagerup-like tensor products with norm bounds at most a constant times ||φ||.
    This nontrivial technical fact, quoted from [ANP] in Section 3, underlies the proof of Theorems 3.1 and 3.2 and is not reproved here.
  • domain assumption Inequalities (2.3) and (2.5) hold: triple operator integrals of the first and second gender with symbols in Haagerup-like tensor products map S_p into S_p with the stated norm bounds for 1 ≤ p ≤ 2.
    These are the engine of the positive results, taken from [ANP] and [AP3]; if they were false, Theorem 3.1 would not follow.
  • domain assumption From §8 of [ANP]: for 2 < p ≤ ∞, there is no Lipschitz estimate ||φ(A1,B) - φ(A2,B)||_{S_p} ≤ K ||φ|| ||A1 - A2||_{S_p} for all finite rank self-adjoint operators.
    This prior theorem is the basis for the negative results in Section 4 and is accepted without proof.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Commutator estimates for functions of noncommuting self-adjoint operators." pith.science (2026). https://pith.science/paper/ZNHBLZPW

@misc{pith2026260816731,
  author       = {Pith},
  title        = {Pith review of: Commutator estimates for functions of noncommuting self-adjoint operators},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZNHBLZPW}},
  note         = {Machine review of arXiv:2608.16731}
}
abstract

We study properties of the calculus $\varphi\mapsto\varphi(A,B)$ for self-adjoint operators with commutator $[A,B]$ in the Schaten--von Neumann class $\boldsymbol{S}_p$. It turns out that this calculus defined on the Besov class $B_{\infty,1}^1({\Bbb R}^2)$ in the case $p\le2$ admits a commutator Lipschitz estimate and is multiplicative modulo $\boldsymbol{S}_p$. On the other hand in the case $p>2$ there is no commutator Lipschitz estimate. There is no commutator Lipschitz estimate in the operator norm as well.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 32 canonical work pages

  1. [1]

    Aleksandrov, F.L

    A.B. Aleksandrov, F.L. Nazarov and V.V. Peller , Functions of noncommuting self-adjoint operators under perturbation and estimates of triple operator integrals , Advances in Math. 295 (2016), 1--52

    work page 2016

  2. [2]

    Aleksandrov A. B. , Peller V. V. , Operator Lipschitz functions , Uspekhi Matem. Nauk. 71:4 (2016), 3--106 (Russian). English transl.: Russian Math. Surveys, 71:4 (2016), 605--702

    work page 2016

  3. [3]

    Aleksandrov and V.V

    A.B. Aleksandrov and V.V. Peller , Functions of almost commuting operators and an extension of the Helton--Howe trace formula , J. Funct. Anal. 271 (2016), 3300--3322

    work page 2016

  4. [4]

    Aleksandrov and V.V

    A.B. Aleksandrov and V.V. Peller , Multiple operator integrals, Haagerup and Haagerup-like tensor products, and operator ideals , Bull. London Math. Soc. 49 (2017), 463--479

    work page 2017

  5. [5]

    Aleksandrov and V.V

    A.B. Aleksandrov and V.V. Peller , Haagerup tensor products and Schur multipliers , Algebra i Analiz 36:5 (2024), 70--85

    2024

  6. [6]

    B., Peller V

    Aleksandrov A. B., Peller V. V. , Schur multipliers of Schatten-von Neumann classes

    work page

  7. [7]

    , Schur multipliers , Duke Math

    Bennett G. , Schur multipliers , Duke Math. J. 44 (1977), 603--639

    work page 1977

  8. [8]

    S., Solomyak M

    Birman M. S., Solomyak M. Z. , Double Stieltjes operator integrals , Problems of Math. Phys., Leningrad. Univ. 1 (1966), 33--67 (Russian). English transl., Topics Math. Physics 1, 25--54 . Consultants Bureau Plenum Publishing Corporation, New York, 1967

    work page 1966

Show all 35 references
  1. [9]

    S., Solomyak M

    Birman M. S., Solomyak M. Z. , Double Stieltjes operator integrals. II , Problems of Math. Phys., Leningrad. Univ. 2 (1967), 26--60 (Russian). English transl., Topics Math. Physics 2, 19--46. Consultants Bureau Plenum Publishing Corporation, New York, 1968

    1967

  2. [10]

    S., Solomyak M

    Birman M. S., Solomyak M. Z. , Double Stieltjes operator integrals. III , Problems of Math. Phys. 6 , Leningrad. Univ, 27--53 (1973) (Russian)

    1973

  3. [11]

    S., Solomyak M

    Birman M. S., Solomyak M. Z. ,

  4. [12]

    Daletskii Yu. L, S. G. Krein ,

  5. [13]

    Gohberg and M.G

    I.C. Gohberg and M.G. Krein, Introduction to the theory of linear nonselfadjoint operators in Hilbert space, Nauka, Moscow, 1965

    1965

  6. [14]

    Helton and R

    J.W. Helton and R. Howe , Integral operators, commutators, traces, index, and homology , in ``Lecture Notes in Math.'', vol. 345 , pp. 141--209, Springer-Verlag, New York, 1973

    1973

  7. [15]

    Juschenko, I.G

    K. Juschenko, I.G. Todorov and L. Turowska , Multidimensional operator multipliers , Trans. Amer. Math. Soc. 361 (2009), 4683--4720

    2009

  8. [16]

    Peller V. V. , Hankel operators in the perturbation theory of unitary and self-adjoint operators , Funk, anal. i ego pril. 19 :2 (1985), 37--51 (in Russian); English transl., Funct. Anal. Appl., 19 :2 (1985), 111--123

    1985

  9. [17]

    , Similarity problem and completely bounded maps ,

    Pisier G. , Similarity problem and completely bounded maps ,

  10. [18]

    ankten Bilinearformen mit unendlich vielen Ver\

    Schur I. , Bemerkungen zur Theorie der beschr\" ankten Bilinearformen mit unendlich vielen Ver\"andlichen , J. Reine Ungew. Math. 140 (1911), 1--28

    1911

  11. [19]

    Peller , Multiple operator integrals and higher operator derivatives , J

    V.V. Peller , Multiple operator integrals and higher operator derivatives , J. Funct. Anal. 233 (2006), 515--544

    2006

  12. [20]

    Peller , Multiple operator integrals in perturbation theory , Bull

    V.V. Peller , Multiple operator integrals in perturbation theory , Bull. Math. Sci. 6 (2016), 15--88

    2016

  13. [21]

    Peller , Besov spaces in operator theory , Russian Math

    V.V. Peller , Besov spaces in operator theory , Russian Math. Surveys 79:1 , 1--52

  14. [22]

    Pincus , Commutators and systems of singular integral equations, I , Acta Math

    J.D. Pincus , Commutators and systems of singular integral equations, I , Acta Math. 121 (1968), 219-249

    1968

  15. [23]

    Aleksandrov and V.V

    A.B. Aleksandrov and V.V. Peller , Estimates of operator moduli of continuity ,

  16. [24]

    Aleksandrov and V.V

    A.B. Aleksandrov and V.V. Peller , Schur multipliers multipliers of Schatten--von Neumann classes _p , J. Funct. Anal. 279 (2020), 108683

    2020

  17. [25]

    Bennett , Schur multipliers , Duke Math

    G. Bennett , Schur multipliers , Duke Math. J. 44 (1977), 603--639

    1977

  18. [26]

    Birman and M.Z

    M.S. Birman and M.Z. Solomyak ,

  19. [27]

    , .. , , . , , 1956, . 1, . 81--106

    1956

  20. [28]

    , 19:2 (1985), 37--51

    , , . ., 19:2 (1985), 37--51

    1985

  21. [29]

    Defant and K

    A. Defant and K. Floret , Tensor norms and operator ideals. North-Holland,

  22. [30]

    Grothendieck , R\'esum\'e de la th\'eorie m\'etrique des produits tensoriels

    A. Grothendieck , R\'esum\'e de la th\'eorie m\'etrique des produits tensoriels

  23. [31]

    Kwapie\'n , On operators factorizable through Lp-space

    S. Kwapie\'n , On operators factorizable through Lp-space. Bull. Soc. Math

  24. [32]

    Lindenstrauss and A

    J. Lindenstrauss and A. Pelczy\'nski , Absolutely summing operators in L _p -

  25. [33]

    Pietsch , Operator ideals

    A. Pietsch , Operator ideals. North-Holland Amsterdam 1978

    1978

  26. [34]

    Pisier , Similarity problem and completely bounded maps ,

    G. Pisier , Similarity problem and completely bounded maps ,

  27. [35]

    ankten Bilinearformen mit unendlich vielen Ver\

    I. Schur , Bemerkungen zur Theorie der beschr\" ankten Bilinearformen mit unendlich vielen Ver\"andlichen , J. Reine Ungew. Math. 140 (1911), 1--28

    1911

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.