Pith. sign in

REVIEW 2 major objections 3 minor 32 references

Determinants of Random Unitary Pencils

T0 review · 2 major / 3 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 exact and limiting formulas for Haar averages of products of two unitary-pencil determinants, establishing the full conjecture for scalar and upper-triangular coefficients and a formal version in general.

desk verdict A clean scalar-case formula and an honest conjecture, but the triangular-case proof has a real, load-bearing gap in the content-ratio argument. read the letter →

arxiv 2506.04400 v1 pith:PW7FVEWU submitted 2025-06-04 math.FA math-phmath.MP

Michael T. Jury , George Roman This is my paper
classification math.FAmath-phmath.MP MSC 15B5260B2005E0546E22
keywords randomunitarypencilsHaarmeasurecharacteristicpolynomialsDrury-ArvesonspaceSchur-WeyldualityLittlewood-RichardsoncoefficientscontentratiosSzegőasymptotics
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 asks what happens to the Haar average of a product of two determinants of random unitary pencils, expressions of the form $\det(I_k\otimes I_d + \sum_{j=1}^g X_j\otimes U_j)$ with $U_j$ independent Haar-distributed $d\times d$ unitaries and $X_j$ fixed matrices. In the scalar-coefficient case the authors derive an exact finite-$d$ formula: the integral is a weighted multinomial sum $\sum_{n=0}^d \sum_{|\alpha|=n} c(d,\alpha)\binom{n}{\alpha} x^\alpha y^\alpha$, whose coefficients $c(d,\alpha)$ increase to $1$ as $d\to\infty$, so the limit is the reproducing kernel $1/(1-\langle x,y\rangle)$ of the Drury-Arveson space. They conjecture that for general matrix coefficients the limit is $\det(I_{kk'}-\sum_{j=1}^g X_j\otimes \bar Y_j)^{-1}$ whenever the outer spectral radii of the tuples are $<1$. They prove this conjecture for upper-triangular coefficient tuples—hence for commuting tuples—and they prove that the identity holds in the formal power series sense for arbitrary tuples, with a uniform $L^2$ bound as the missing ingredient.

What carries the argument

The machinery is the expansion of $\det(L_X(U))$ into trace monomials $p_{\sigma,\alpha}(U)=\operatorname{tr}(\rho^d_{S_n}(\sigma^{-1})U^{\otimes\alpha})$ over the symmetric group, with Schur-Weyl duality separating the coefficient matrices $X$ from the unitaries $U$. For the scalar exact computation the load-bearing tools are the conditional expectation $E_\alpha$ onto the range of the Young subgroup $S_\alpha$ and the central projection $Q_\varepsilon$ onto the sign representation of $S_n$. For the triangular theorem the decisive quantity is the content ratio $r^\lambda_{\mu,\nu}(d)=C_\lambda(d)/(C_\mu(d)C_\nu(d))$, where $C_\lambda(d)=\prod_{u\in\lambda}(d+c_u)$ is the content polynomial of a partition; Theorem 5.8 bounds every such ratio by $(n+1)^{k^2}$, and that bound supplies the $d$-uniform $L^2$ control.

What would settle it

Compute $C_\lambda(d)/(C_\mu(d)C_\nu(d))$ for a concrete partition $\lambda$ with at most $d$ rows and largest part at most $k$ and subpartitions $\mu,\nu$ with nonzero Littlewood-Richardson coefficient; a single triple exceeding $(n+1)^{k^2}$ would falsify Theorem 5.8 and break the triangular theorem. At the level of the conjecture, Monte-Carlo integration of the Haar average for a non-triangular $2\times2$ coefficient pair at increasing $d$, compared with the predicted inverse determinant, would settle whether the general formula holds.

Watch

Extended reading notes

Core claim

The central claim is Conjecture 3.1: for $g$-tuples of square matrices $X=(X_1,\dots,X_g)$ and $Y=(Y_1,\dots,Y_g)$ of sizes $k\times k$ and $k'\times k'$ with outer spectral radius $\operatorname{rad}(X),\operatorname{rad}(Y)<1$, the expectation over $U(d)^g$ of $\det(L_X(U))\,\det(L_Y(U))$ tends, as $d\to\infty$, to $\det(I_k\otimes I_{k'} - \sum_{j=1}^g X_j\otimes \bar Y_j)^{-1}$. The main proved results are Theorem 4.1, an exact formula at every finite $d$ for scalar coefficients, and Theorem 5.1, the conjecture for upper-triangular coefficient tuples. The triangular case is reached by reducing it to coefficients that are scalar multiples of the identity and then proving that the $2k$-th moment of $|\det(I+\sum_j x_j U_j)|$ is bounded uniformly in $d$ for $\|x\|_2<1$. The authors also establish the formal identity by matching homogeneous coefficients in the expansions of both sides, and they show that a general uniform $L^2$ estimate would upgrade the formal identity to convergence.

Load-bearing premise

Everything in the triangular theorem rides on the combinatorial content-ratio bound—that $C_\lambda(d)/(C_\mu(d)C_\nu(d)) \le (n+1)^{k^2}$ for every partition $\lambda$ of height $\le d$ and width $\le k$ and every admissible subpartitions $\mu,\nu$—because if that polynomial bound were too small, the $d$-uniform moment estimates would fail and the proof would collapse.

Editorial extensions

If this is right

  • In the scalar case the finite-$d$ integral is exactly $\sum_{|\alpha|\le d} c(d,\alpha)\binom{n}{\alpha} x^\alpha y^\alpha$, and its limit is $1/(1-\langle x,y\rangle)$, the Drury-Arveson kernel (Corollary 4.2).
  • For upper-triangular, diagonal, or commuting coefficient tuples the limit factors as $\prod_{l=1}^k\prod_{m=1}^{k'}(1-\langle x_l,y_m\rangle)^{-1}$, so the conjecture holds in those cases (Corollaries 1.4, 1.5, 5.2).
  • The identity holds in the formal-power-series sense for all tuples with outer spectral radius $<1$, and a uniform $L^2$ bound on the determinant would turn it into actual convergence (Corollary 3.4, Proposition 3.5).
  • If the conjecture holds, the limit extends to determinants of stable noncommutative polynomials in the unitaries, with value $\det(I_{kk'}-\sum_j X_j\otimes \bar Y_j)^{-1}$ for a linearizing pair of tuples (Proposition 7.1).
  • Conic linear combinations of independent Haar unitaries have characteristic-polynomial moments with Szegő-type asymptotics, and the boundary of the asymptotic regime is the Brown-measure disk of radius $\sqrt{g}$ (Corollary 7.2).

Reading between the lines

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

  • A natural test of the full conjecture is a numerical Haar integral for small non-triangular, noncommuting coefficient tuples (for instance $g=2$ and $2\times2$ nilpotent matrices), where formal coefficient matching holds but the $d$-uniform bounds are unknown.
  • The explicit finite-$d$ coefficients $c(d,\alpha)$ are ratios of falling factorials that approach $1$; extracting their $1/d$ expansion could yield refined Szegő-type asymptotics for matrix-coefficient pencils.
  • The appearance of the kernel $(1-\langle x,y\rangle)^{-1}$ suggests the limiting average defines a reproducing kernel on the unit ball of row contractions, which could connect these random-matrix averages to multivariable operator-theoretic interpolation theorems.
  • The content-ratio bound has the flavor of log-concavity for content polynomials; a sharper or more structural proof of $C_\lambda(d) \le C_\mu(d)C_\nu(d)$ times a fixed polynomial factor would likely deliver the full conjecture via Proposition 3.5.
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 studies Haar expectations over U(d)^g of products det(L_X(U)) det(conj(L_Y(U))) for unitary pencils L_X(U)=I_k⊗I_d+∑_{j=1}^g X_j⊗U_j. The central object is the large-d limit, conjectured in (1.9)/(3.1) to equal det(I⊗I−∑ X_j⊗conj(Y_j))^{-1} when the outer spectral radii are less than 1. The paper proves an exact finite-d formula in the scalar-coefficient case (Theorem 4.1), establishes formal convergence by matching homogeneous expansions (Section 3), and claims a proof of the conjecture for upper-triangular coefficient tuples (Theorem 5.1) via a uniform L^2-moment bound. The burden of the triangular case falls on a combinatorial content-ratio estimate, Theorem 5.8, whose proof occupies Section 6.

Significance. If the conjecture and the triangular theorem were fully established, this would be a substantial contribution to random matrix theory and multivariable operator theory: the scalar limit identifies the Drury-Arveson kernel as the large-d limit, and the triangular case yields Szegő-type asymptotics for moments of characteristic polynomials of sums of independent Haar unitaries, with the outer spectral radius as the natural convergence domain. The paper is also commendable for its clear use of Schur-Weyl duality, Rădulescu's asymptotic orthogonality, and the normal-families reduction, and the scalar-coefficient theorem appears correct and carefully proved. The advertised triangular theorem, however, currently depends on a combinatorial estimate whose proof contains a false dominance-order statement; until that estimate is properly proved, the main new theorem beyond scalar coefficients is not established.

major comments (2)
  1. [Section 6, Corollary 6.2; used in Proposition 6.3 and Theorem 5.8] Corollary 6.2 asserts that if c^λ_{μν}≠0 then rows(λ/μ)⊴ν and consequently r^λ_{μν}(d)≤r^λ_{μ,rows(λ/μ)}(d). The dominance direction is reversed: McNamara's necessary condition is ν⊴rows(λ/μ), not rows(λ/μ)⊴ν. The printed statement is falsified by λ=(2,2), d=2, μ=(2), ν=(1,1): here c^{(2,2)}_{(2),(1,1)}=1, C_λ(2)=12, C_μ(2)=6, C_ν(2)=2, so r^λ_{μν}(2)=1, while rows(λ/μ)=(2) gives C_rows(2)=6 and r^λ_{μ,rows}(2)=1/3. Thus the asserted inequality fails. Since Proposition 6.3 uses Corollary 6.2 to reduce to the case ν=rows(λ/μ), and Theorem 5.8 is the key input for the uniform bound (5.1) and therefore for Theorem 5.1, the proof of the triangular case as printed is incomplete. The final bound in Theorem 5.8 may be true, but the argument supplied does not establish it.
  2. [Section 5.1, Theorem 5.11] Even assuming Theorem 5.8, the proof of Theorem 5.11 for g>2 applies Theorem 5.8 to each factor r^{ν_{i-1}}_{μ_i,ν_i}, but the outer partition ν_{i-1} has size |ν_{i-1}|=α_i+...+α_g, which may be strictly smaller than k even when the ambient n=|α| satisfies n≥k. Theorem 5.8 is stated only under the hypothesis k≤n≤kd, so these applications are outside the stated hypotheses. A separate bound for subpartitions of size below k is needed; as written, the claimed extension to g>2 is not justified even after repairing Theorem 5.8.
minor comments (3)
  1. [Section 1.1] The word 'indpendently' should be 'independently'.
  2. [Equation (5.8)] The displayed formula is missing the inner summation over multi-indices α of weight n; the text should read ∑_{n=0}^{kd} ∑_{|α|=n}, as is used later in the proof of Theorem 5.1.
  3. [Section 6, Proposition 6.3 example] The numerical example after Proposition 6.3 is helpful, but it also illustrates that the update procedure does not strictly increase the content ratio at every step: starting from λ=(4,4,3), μ=(3,2,0), Update (A) gives a smaller ratio and Update (B) gives a larger one, yet neither updated pair satisfies λ=μ+ν. The example is therefore consistent with the repair needed, but the surrounding text should make clear that the proof requires a full argument rather than a single successful update.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the central derivation is self-contained, with only a minor self-citation in an auxiliary conditional application.

full rationale

The paper's derivation chain is not circular. The main results are obtained by expanding det(L_X(U)) and the proposed limit into homogeneous trace-monomial series (Lemmas 3.1 and 3.2), applying Rădulescu's asymptotic orthogonality as an external input (Theorem 3.3), and showing termwise coefficient matching (Corollary 3.4). The scalar case is an exact finite-size computation (Theorem 4.1), while the triangular case is reduced to uniform L^2 bounds (Proposition 5.3) and a combinatorial estimate for content ratios (Theorem 5.8). None of these steps assumes Conjecture (1.9); the conjectured determinant identity is the output of the calculation, not an input. There are no fitted parameters and no quantity is renamed as a prediction. The only self-citation by an author appears in the later, conditional application Proposition 7.1, which invokes [15] for a stability/realization theorem and asserts without proof that it extends to matrix coefficients; this does not carry the central claim and is not used to prove Theorem 1.1 or Theorem 1.3. Separately, the proof of Theorem 5.8 via Corollary 6.2 and Proposition 6.3 appears fragile, and the dominance-order direction in Corollary 6.2 may be misstated; however, a gap or error in an estimate is a correctness concern, not circularity. Accordingly, no circular step is identified, and the score 2 reflects only the minor, non-load-bearing self-citation in the auxiliary application.

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

The central theorems are derived from published results, namely Schur-Weyl duality, Rădulescu asymptotics, Littlewood-Richardson theory, and Weingarten calculus, rather than from new postulates. There are no fitted free parameters. The only internal estimate that goes beyond bookkeeping is Theorem 5.8, whose proof as printed contains a repairable typo in Proposition 6.3.

assumptions (6)
  • standard math Schur-Weyl duality between the symmetric group S_n and U(d) acting on (C^d)⊗n
    Used in Section 2.5, Eq. (2.15), to decompose tensor powers and in Propositions 4.8, 5.4, and 5.5.
  • domain assumption Rădulescu's asymptotic orthogonality theorem (Theorem 4.1 of the cited paper [26])
    Invoked as Theorem 3.3 to compute the d→∞ limit of integrals of trace monomials; the constants in the limit are load-bearing for Corollary 3.4 and the formal convergence.
  • standard math McNamara's dominance result: rows(λ/μ)⊴ν for every ν with c^λ_{μ,ν}≠0, and c^λ_{μ,rows(λ/μ)}≠0
    Used in Corollary 6.2 to replace an arbitrary ν by rows(λ/μ) when maximizing content ratios.
  • standard math Stanley hook-content formulas for Schur polynomials and symmetric group characters
    Used throughout Sections 4-6, especially (2.10), (2.14), and Lemma 5.6, to relate Schur evaluations, character dimensions, and content polynomials.
  • standard math Collins-Śniady Weingarten calculus, in particular the projection property of E_j and the identity Φ_j = Φ_j(I)E_j
    Used in Lemmas 4.5-4.7 and Corollary 4.7 to express E_α as a projection onto the group algebra of the Young subgroup S_α.
  • domain assumption Popescu/Pascoe characterization: a tuple has outer spectral radius below 1 if and only if it is jointly similar to a row contraction
    Used in Proposition 2.5 and Proposition 2.6 to pass between spectral-radius assumptions and contraction assumptions in the normal-families arguments.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Determinants of Random Unitary Pencils." pith.science (2026). https://pith.science/paper/PW7FVEWU

@misc{pith2026250604400,
  author       = {Pith},
  title        = {Pith review of: Determinants of Random Unitary Pencils},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PW7FVEWU}},
  note         = {Machine review of arXiv:2506.04400}
}
read the original abstract

We investigate determinants of random unitary pencils (with scalar or matrix coefficients), which generalize the characteristic polynomial of a single unitary matrix. In particular we examine moments of such determinants, obtained by integrating against the Haar measure on the unitary group. We obtain an exact formula in the case of scalar coefficients, and conjecture an asymptotic formula in the general case, and prove a special case of the conjecture.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

32 extracted references · 30 canonical work pages

  1. [1]

    Jinho Baik and Eric M. Rains. Algebraic aspects of increasing subsequences.Duke Math. J., 109(1):1–65, 2001

    work page 2001

  2. [2]

    Limiting spectral distribution of sums of unitary and orthogonal matrices.Electron

    Anirban Basak and Amir Dembo. Limiting spectral distribution of sums of unitary and orthogonal matrices.Electron. Commun. Probab., 18:no. 69, 19, 2013

    2013

  3. [3]

    Factoring determinants and applications to number theory

    Estelle Basor and Brian Conrey. Factoring determinants and applications to number theory. Random Matrices Theory Appl., 13(2):Paper No. 2450010, 33, 2024

    work page 2024

  4. [4]

    Borodin and E

    A. Borodin and E. Strahov. Averages of characteristic polynomials in random matrix theory. Comm. Pure Appl. Math., 59(2):161–253, 2006

    work page 2006

  5. [5]

    Springer-Verlag, New York, 2004

    Daniel Bump.Lie groups, volume 225 ofGraduate Texts in Mathematics. Springer-Verlag, New York, 2004

    work page 2004

  6. [6]

    Toeplitz minors.J

    Daniel Bump and Persi Diaconis. Toeplitz minors.J. Combin. Theory Ser. A, 97(2):252–271, 2002

    work page 2002

  7. [7]

    On the averages of characteristic polynomials from classical groups.Comm

    Daniel Bump and Alex Gamburd. On the averages of characteristic polynomials from classical groups.Comm. Math. Phys., 265(1):227–274, 2006

    work page 2006

  8. [8]

    Moments and cumulants of polynomial random variables on unitary groups, the Itzykson-Zuber integral, and free probability.Int

    Beno ˆ ıt Collins. Moments and cumulants of polynomial random variables on unitary groups, the Itzykson-Zuber integral, and free probability.Int. Math. Res. Not., (17):953–982, 2003

    work page 2003

Show all 32 references
  1. [9]

    Integration with respect to the Haar measure on unitary, orthogonal and symplectic group.Comm

    Beno ˆ ıt Collins and Piotr´Sniady. Integration with respect to the Haar measure on unitary, orthogonal and symplectic group.Comm. Math. Phys., 264(3):773–795, 2006

    2006

  2. [10]

    J. B. Conrey, D. W. Farmer, J. P. Keating, M. O. Rubinstein, and N. C. Snaith. Autocorre- lation of random matrix polynomials.Comm. Math. Phys., 237(3):365–395, 2003

    2003

  3. [11]

    Springer-Verlag, New York, 1991

    William Fulton and Joe Harris.Representation theory, volume 129 ofGraduate Texts in Mathematics. Springer-Verlag, New York, 1991

    1991

  4. [12]

    Brown’s spectral distribution measure forR-diagonal elements in finite von Neumann algebras.J

    Uffe Haagerup and Flemming Larsen. Brown’s spectral distribution measure forR-diagonal elements in finite von Neumann algebras.J. Funct. Anal., 176(2):331–367, 2000

    2000

  5. [13]

    An invitation to the Drury-Arveson space

    Michael Hartz. An invitation to the Drury-Arveson space. InLectures on analytic function spaces and their applications, volume 39 ofFields Inst. Monogr., pages 347–413. Springer, Cham, 2023. DETERMINANTS OF RANDOM UNITARY PENCILS 49

    2023

  6. [14]

    William Helton, Igor Klep, and Jurij Volˇ ciˇ c

    J. William Helton, Igor Klep, and Jurij Volˇ ciˇ c. Geometry of free loci and factorization of noncommutative polynomials.Adv. Math., 331:589–626, 2018

    2018

  7. [15]

    Jury, Robert T

    Michael T. Jury, Robert T. W. Martin, and Eli Shamovich. Non-commutative rational func- tions in the full Fock space.Trans. Amer. Math. Soc., 374(9):6727–6749, 2021

    2021

  8. [16]

    J. P. Keating and N. C. Snaith. Random matrix theory andζ(1/2 +it).Comm. Math. Phys., 214(1):57–89, 2000

    2000

  9. [17]

    A theorem of Frobenius, a theorem of Amitsur-Levitski and cohomology theory.J

    Bertram Kostant. A theorem of Frobenius, a theorem of Amitsur-Levitski and cohomology theory.J. Math. Mech., 7:237–264, 1958

    1958

  10. [18]

    I. G. Macdonald.Symmetric functions and Hall polynomials. Oxford Classic Texts in the Physical Sciences. The Clarendon Press, Oxford University Press, New York, second edition, 2015

    2015

  11. [19]

    Matrix group integrals, surfaces, and mapping class groups I:U(n).Invent

    Michael Magee and Doron Puder. Matrix group integrals, surfaces, and mapping class groups I:U(n).Invent. Math., 218(2):341–411, 2019

    2019

  12. [20]

    Peter R. W. McNamara. Necessary conditions for Schur-positivity.J. Algebraic Combin., 28(4):495–507, 2008

    2008

  13. [21]

    Mingo, Piotr ´Sniady, and Roland Speicher

    James A. Mingo, Piotr ´Sniady, and Roland Speicher. Second order freeness and fluctuations of random matrices. II. Unitary random matrices.Adv. Math., 209(1):212–240, 2007

    2007

  14. [22]

    James E. Pascoe. The outer spectral radius and dynamics of completely positive maps.Israel J. Math., 244(2):945–969, 2021

    2021

  15. [23]

    Similarity problems in noncommutative polydomains.J

    Gelu Popescu. Similarity problems in noncommutative polydomains.J. Funct. Anal., 267(11):4446–4498, 2014

    2014

  16. [24]

    C. Procesi. The invariant theory ofn×nmatrices.Advances in Math., 19(3):306–381, 1976

    1976

  17. [25]

    A note on the Formanek Weingarten function.Note Mat., 41(1):69–109, 2021

    Claudio Procesi. A note on the Formanek Weingarten function.Note Mat., 41(1):69–109, 2021

    2021

  18. [26]

    Combinatorial aspects of Connes’s embedding conjecture and asymptotic distribution of traces of products of unitaries

    Florin R˘ adulescu. Combinatorial aspects of Connes’s embedding conjecture and asymptotic distribution of traces of products of unitaries. InOperator theory 20, volume 6 ofTheta Ser. Adv. Math., pages 197–205. Theta, Bucharest, 2006

    2006

  19. [27]

    Sagan.The symmetric group, volume 203 ofGraduate Texts in Mathematics

    Bruce E. Sagan.The symmetric group, volume 203 ofGraduate Texts in Mathematics. Springer-Verlag, New York, second edition, 2001

    2001

  20. [28]

    Graduate Texts in Mathematics, Vol

    Jean-Pierre Serre.Linear representations of finite groups. Graduate Texts in Mathematics, Vol. 42. Springer-Verlag, New York-Heidelberg, 1977. Translated from the second French edition by Leonard L. Scott

    1977

  21. [29]

    Stanley.Enumerative combinatorics

    Richard P. Stanley.Enumerative combinatorics. Volume 1, volume 49 ofCambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, second edition, 2012

    2012

  22. [30]

    Stanley.Enumerative combinatorics

    Richard P. Stanley.Enumerative combinatorics. Vol. 2, volume 208 ofCambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2024. Second edition

    2024

  23. [31]

    On domains of noncommutative rational functions.Linear Algebra Appl., 516:69– 81, 2017

    Jurij Volˇ ciˇ c. On domains of noncommutative rational functions.Linear Algebra Appl., 516:69– 81, 2017

    2017

  24. [32]

    Princeton Landmarks in Mathematics

    Hermann Weyl.The classical groups. Princeton Landmarks in Mathematics. Princeton Uni- versity Press, Princeton, NJ, 1997. Their invariants and representations, Fifteenth printing, Princeton Paperbacks. Department of Mathematics, University of Florida, PO Box 118105, Gainesvill...

    1997

Pith tools

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