pith. sign in

arxiv: 2510.03382 · v5 · pith:M6RWMEXSnew · submitted 2025-10-03 · 🧮 math.OA · math-ph· math.MP· math.PR

Spectral results for free random variables

Brian C. Hall , Ching-Wei Ho This is my paper

Pith reviewed 2026-05-21 20:58 UTC · model grok-4.3

classification 🧮 math.OA math-phmath.MPmath.PR
keywords Brown measureoperator spectrumfree random variablesvon Neumann algebraanalytic extensioncircular elementselliptic elementsmultiplicative Brownian motion
0
0 comments X

The pith

If the derivative of the trace-log function extends analytically near zero, then lambda lies outside the spectrum of a.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes a criterion for locating the spectrum of an element a in a von Neumann algebra equipped with a faithful normal trace. If the partial derivative with respect to a small positive epsilon of the trace of the logarithm of ((a minus lambda) adjoint times (a minus lambda) plus epsilon) admits a real analytic extension to a neighborhood of zero, then lambda does not belong to the spectrum of a. A reader would care because the criterion supplies a practical test for excluding points from the spectrum in free probability. The authors apply the criterion to circular elements, elliptic elements, and free multiplicative Brownian motions, concluding in most cases that the spectrum equals the support of the Brown measure.

Core claim

Suppose that for a fixed λ in the complex numbers, the function ε mapping to the trace of the inverse of ((a−λ)∗(a−λ)+ε) admits a real analytic extension to a neighborhood of 0 in the reals. Then λ is outside the spectrum of a. The result is applied to circular and elliptic elements as well as free multiplicative Brownian motions, showing that the spectrum coincides with the support of the Brown measure in most cases.

What carries the argument

The real analytic extension near zero of the ε-derivative of S, equal to the trace of ((a−λ)∗(a−λ)+ε)^{-1}, which functions as the test for lambda lying in the resolvent set.

If this is right

  • The spectrum of circular elements equals the support of their Brown measure.
  • The spectrum of elliptic elements equals the support of their Brown measure.
  • The spectrum of free multiplicative Brownian motions equals the support of their Brown measure in most cases.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The analyticity test could be checked on additional classes of free random variables beyond the three families treated in the paper.
  • The same condition might connect to questions about spectra in random matrix ensembles that approximate free elements.
  • One could ask whether the criterion extends to settings with non-normal traces or to elements in other algebras.

Load-bearing premise

The partial derivative with respect to epsilon of the trace of the inverse must admit a real analytic extension to a neighborhood of zero.

What would settle it

An explicit operator a together with a lambda for which the derivative extends analytically to zero yet lambda still lies in the spectrum of a would disprove the implication.

Figures

Figures reproduced from arXiv: 2510.03382 by Brian C. Hall, Ching-Wei Ho.

Figure 1
Figure 1. Figure 1: The region Σt from [9] for t = 4 (left), and a detail thereof (right). We now briefly summarize the applications we will make of Theorem 2.6. We consider a element of the form x+c, where c is circular and x is freely independent of x, or more generally x + g, where g is elliptic and freely independent of x. Assume that the spectrum of x coincides with the support of its Brown measure, which will happen, fo… view at source ↗
Figure 2
Figure 2. Figure 2: Density plot of the Brown measure of x + ct, where x is self-adjoint with µx = 1 2 (δ−1 + δ1), with t = 1. elliptic elements as “additive,” since the sum of two freely independent elliptic elements is again elliptic. Specifically, if gt1,γ1 and gt2,γ2 are freely independent elliptic elements, then gt1,γ1 + gt2,γ2 d= gt1+t2,γ1+γ2 , (3.6) where d= denotes equality in ∗-distribution. (Compare (3.13) in the mu… view at source ↗
Figure 4
Figure 4. Figure 4: Finally, Zhong [34] computed the Brown measure of [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 3
Figure 3. Figure 3: Density plot of the Brown measure of x+ct, where x is unitary and µx is supported at the third roots of unity, with equal masses, and t = 1 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Density plot of the Brown measure of x + gt,γ with x self-adjoint and µx = 1 2 (δ−1 + δ1), with t = 1 and γ = −i/2. Let x = uh, where u is a Haar unitary that is freely independent of h. Then by Ex￾ample 2.2, the spectrum of x is a proper subset of its Brown support. Furthermore, for all sufficiently small t, the spectrum of x + ct is a proper subset of its Brown support. The proof of this result is deferr… view at source ↗
Figure 5
Figure 5. Figure 5: Density plot of Brown measure of ubt, where u is uni￾tary with µu supported at the fourth roots of unity, with equal mass, for t = 1 [PITH_FULL_IMAGE:figures/full_fig_p015_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Density plot of Brown measure of ubt,γ, where u is unitary with µu supported at the fourth roots of unity, with equal mass, for t = 1 and γ = −i/2 [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The domains Σt (left) and Dt,γ (right) for t = γ = 2. spectrum of xbt,γ. Furthermore, if 0 is outside Σt, then the point 0 = fγ(0) is in the spectrum of xbt,γ if and only if µx({0}) > 0. Proof. The analysis of 0 is similar to the proof of Theorem 5.9, using [10, Propo￾sition 4.13] in place of [10, Theorem 3.21]. If λ is a nonzero point outside Σt and we assume T(λ) > 0, then Theorem 5.9 tells us that λ is … view at source ↗
read the original abstract

Let $(\mathcal{A},\mathrm{tr})$ be a von Neumann algebra with a faithful, normal trace $\mathrm{tr}:\mathcal{A}\rightarrow\mathbb{C}.$ For each $a\in\mathcal{A},$ define \[ S(\lambda,\varepsilon)=\mathrm{tr}[\log((a-\lambda)^{\ast}(a-\lambda )+\varepsilon)],\quad\lambda\in\mathbb{C},~\varepsilon>0, \] so that the limit as $\varepsilon\rightarrow0^{+}$ of $S$ is the log potential of the Brown measure of $a.$ Suppose that for a fixed $\lambda\in\mathbb{C},$ the function \[ \varepsilon\mapsto\frac{\partial S}{\partial\varepsilon}(\lambda ,\varepsilon)=\mathrm{tr}[((a-\lambda)^{\ast}(a-\lambda)+\varepsilon )^{-1}] \] admits a real analytic extension to a neighborhood of $0$ in $\mathbb{R}.$ Then we will show that $\lambda$ is outside the spectrum of $a.$ We will apply this result to several examples involving circular and elliptic elements, as well as free multiplicative Brownian motions. In most cases, we will show that the spectrum of the relevant element $a$ coincides with the support of its Brown measure.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that in a von Neumann algebra (A, tr) equipped with a faithful normal trace, if the map ε ↦ tr[((a−λ)^*(a−λ) + ε)^{-1}] admits a real-analytic extension to a neighborhood of 0 in R, then λ lies outside the spectrum of a. The authors derive this from properties of the Stieltjes transform of the spectral measure of b = (a−λ)^*(a−λ) and apply the criterion to circular elements, elliptic elements, and free multiplicative Brownian motions, concluding that the spectrum coincides with the support of the Brown measure in most cases.

Significance. If the central implication holds, the result supplies a concrete analytic test (via real-analytic continuation of the resolvent trace) for locating the spectrum relative to the Brown measure. This is useful in free probability for establishing spectrum-support equality without direct invertibility arguments, and the reduction to the Stieltjes transform of a positive operator is a clear strength that makes the criterion falsifiable in concrete examples.

major comments (2)
  1. [Main result / proof of the implication from analyticity to invertibility] The main result (stated in the abstract and presumably proved in §2) reduces the claim to the standard fact that real-analyticity of the Stieltjes transform on (−δ,δ) forces the spectral measure μ of b to satisfy supp(μ) ∩ [0,δ) = ∅. The manuscript should explicitly recall or cite the precise complex-analysis statement used (e.g., identity theorem or analytic continuation across a cut), as this step is load-bearing for the implication that b ≥ δI and hence a−λ is invertible.
  2. [Applications section (Brownian motion example)] In the applications to free multiplicative Brownian motion (presumably §4 or the final section), the verification that ε ↦ tr[((a−λ)^*(a−λ)+ε)^{-1}] extends analytically across 0 must be checked against the explicit form of the Brown measure or the S-transform; if the extension is only shown for |λ| larger than the radius of the support, the claim that spectrum equals support of the Brown measure rests on this computation and requires an explicit formula or reference to the relevant free-probability calculation.
minor comments (2)
  1. [Abstract] The definition of S(λ,ε) in the abstract contains a minor LaTeX spacing inconsistency: (a-λ)∗(a-λ )+ε should be written uniformly without the extra space before the closing parenthesis.
  2. [Introduction / notation paragraph] Notation for the partial derivative ∂S/∂ε should be introduced once with a clear statement that the trace is finite and normal, to avoid any ambiguity when the algebra is infinite-dimensional.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive suggestions. The comments help clarify the main implication and strengthen the applications. We address each point below.

read point-by-point responses

  1. Referee: [Main result / proof of the implication from analyticity to invertibility] The main result (stated in the abstract and presumably proved in §2) reduces the claim to the standard fact that real-analyticity of the Stieltjes transform on (−δ,δ) forces the spectral measure μ of b to satisfy supp(μ) ∩ [0,δ) = ∅. The manuscript should explicitly recall or cite the precise complex-analysis statement used (e.g., identity theorem or analytic continuation across a cut), as this step is load-bearing for the implication that b ≥ δI and hence a−λ is invertible.

    Authors: We agree that the load-bearing step benefits from an explicit reference. The argument in §2 uses the fact that a real-analytic Stieltjes transform on an interval around zero implies the positive measure has empty support there, which follows from the identity theorem (the transform extends holomorphically across the cut only if the measure vanishes). In the revised manuscript we will state this explicitly and cite a standard reference such as Theorem 2.4 in Ransford's Potential Theory in the Complex Plane or the relevant uniqueness result for Stieltjes transforms. revision: yes

  2. Referee: [Applications section (Brownian motion example)] In the applications to free multiplicative Brownian motion (presumably §4 or the final section), the verification that ε ↦ tr[((a−λ)^*(a−λ)+ε)^{-1}] extends analytically across 0 must be checked against the explicit form of the Brown measure or the S-transform; if the extension is only shown for |λ| larger than the radius of the support, the claim that spectrum equals support of the Brown measure rests on this computation and requires an explicit formula or reference to the relevant free-probability calculation.

    Authors: The Brown measure of the free multiplicative Brownian motion is known explicitly via the S-transform (see e.g. the computations in the literature on free stochastic processes). For |λ| outside the support the resolvent trace extends analytically by direct differentiation of the known potential; inside the support the singularity is present. In the revision we will add the precise reference to the Brown-measure formula and a short paragraph confirming that the analytic extension holds exactly outside the support, thereby justifying the equality of spectrum and Brown-measure support. revision: yes

Circularity Check

0 steps flagged

No significant circularity in the derivation

full rationale

The paper's central implication is established by identifying the given trace expression as the Stieltjes transform of the spectral measure of the positive operator b = (a − λ)^*(a − λ). Real-analytic extension across zero then forces the support of this measure to be bounded away from zero, implying b ≥ δI for some δ > 0 and hence invertibility of a − λ in the finite von Neumann algebra. This reasoning rests on standard facts about traces, resolvents, and spectral measures rather than any self-referential definitions, fitted parameters renamed as predictions, or load-bearing self-citations. The applications to circular, elliptic, and Brownian-motion elements invoke the same general result without reducing the claim to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper relies on standard axioms from operator algebra theory and analytic function theory; no free parameters or new entities are introduced in the abstract.

axioms (1)
  • domain assumption The existence of a von Neumann algebra with a faithful normal trace tr.
    This is the setting for defining S and the Brown measure.

pith-pipeline@v0.9.0 · 5743 in / 1404 out tokens · 97058 ms · 2026-05-21T20:58:08.475347+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages

  1. [1]

    Belinschi, H

    S. Belinschi, H. Bercovici, and C.-W. Ho, On the support of free convolutions, arXiv:2408.06573 [math.OA]

    work page arXiv

  2. [2]

    Biane, Free Brownian motion, free stochastic calculus and random matrices,Fields Inst

    P. Biane, Free Brownian motion, free stochastic calculus and random matrices,Fields Inst. Commun.,12American Mathematical Society, 1997, 1–19

    work page 1997

  3. [3]

    Biane, On the free convolution with a semi-circular distribution,Indiana Univ

    P. Biane, On the free convolution with a semi-circular distribution,Indiana Univ. Math. J. 46(1997), 705–718

    work page 1997

  4. [4]

    Biane, Segal-Bargmann transform, functional calculus on matrix spaces and the theory of semi-circular and circular systems,J

    P. Biane, Segal-Bargmann transform, functional calculus on matrix spaces and the theory of semi-circular and circular systems,J. Funct. Anal.144(1997), 232–286

    work page 1997

  5. [5]

    Biane and F

    P. Biane and F. Lehner, Computation of some examples of Brown’s spectral measure in free probability,Colloq. Math.90(2001), 181–211

    work page 2001

  6. [6]

    L. G. Brown, Lidski˘ ı’s theorem in the type II case.In: Geometric methods in operator algebras (Kyoto, 1983), Pitman Res. Notes Math. Ser., 123, Longman Scientific & Technical, 1986, 1-35

    work page 1983

  7. [7]

    Capitaine, C

    M. Capitaine, C. Donati-Martin, D. F´ eral, and M. F´ evrier, Free convolution with a semicir- cular distribution and eigenvalues of spiked deformations of Wigner matrices,Electron. J. Probab.16(2011), 64, 1750–1792

    work page 2011

  8. [8]

    Demni and T

    N. Demni and T. Hamdi, Support of the Brown measure of the product of a free unitary Brownian motion by a free self-adjoint projection,J. Funct. Anal.282(2022), Paper No. 109362, 36 pp. 36 BRIAN C. HALL AND CHING-WEI HO

    work page 2022

  9. [9]

    B. K. Driver, B. C. Hall, and T. Kemp, The Brown measure of the free multiplicative Brow- nian motion,Probab. Theory Related Fields184(2022), 209–273

    work page 2022

  10. [10]

    Eaknipitsari and B

    S. Eaknipitsari and B. C. Hall, Support of the Brown measure of a family of free multiplicative Brownian motions with non-negative initial condition, arXiv:2503.19682 [math.PR]

    work page arXiv

  11. [11]

    Erd˝ os and H

    L. Erd˝ os and H. C. Ji, Density of Brown measure of free circular Brownian motion,Doc. Math.30(2025), 417–453

    work page 2025

  12. [12]

    Partial Differential Equations,

    L. Evans, “Partial Differential Equations,” Second edition, Graduate Studies in Mathematics 19, American Mathematical Society, 2010

    work page 2010

  13. [13]

    Gross, Uniqueness of ground states for Schr¨ odinger operators over loop groups,J

    L. Gross, Uniqueness of ground states for Schr¨ odinger operators over loop groups,J. Funct. Anal.112(1993), 373–441

    work page 1993

  14. [14]

    Gross and P

    L. Gross and P. Malliavin, Hall’s transform and the Segal-Bargmann map.In:Itˆ o’s stochastic calculus and probability theory, 73–116. Springer-Verlag, Tokyo, 1996

    work page 1996

  15. [15]

    Haagerup and F

    U. Haagerup and F. Larsen, Brown’s spectral distribution measure for R-diagonal elements in finite von Neumann algebras,J. Funct. Anal.176(2000), 331–367

    work page 2000

  16. [16]

    coherent state

    B. C. Hall, The Segal-Bargmann “coherent state” transform for compact Lie groups,J. Funct. Anal.122(1994), 103–151

    work page 1994

  17. [17]

    B. C. Hall, Quantum theory for mathematicians. Graduate Texts in Mathematics, 267. Springer, New York, 2013. xvi+554 pp

    work page 2013

  18. [18]

    B. C. Hall, PDE methods in random matrix theory.In: Harmonic analysis and applications, Springer Optim. Appl., 168, Springer, 2021, 77-124

    work page 2021

  19. [19]

    B. C. Hall and C.-W. Ho, The Brown measure of the sum of a self-adjoint element and an imaginary multiple of a semicircular element,Lett. Math. Phys.112(2022), Paper No. 19, 61 pp

    work page 2022

  20. [20]

    B. C. Hall and C.-W. Ho, The Brown measure of a family of free multiplicative Brownian motions,Prob. Theory Related Fields,186(2023), 1081–1166

    work page 2023

  21. [21]

    B. C. Hall and C.-W. Ho, The heat flow conjecture for polynomials and random matrices, Lett. Math. Phys.115(2025), Paper No. 60, 53 pp

    work page 2025

  22. [22]

    B. C. Hall, C.-W. Ho, J. Jalowy, and Z. Kabluchko, Zeros of random polynomials undergoing the heat flow, arXiv:2308.11685 [math.PR]

    work page arXiv

  23. [23]

    B. C. Hall and T. Kemp, Brown measure support and the free multiplicative Brownian motion,Adv. Math.355(2019), 106771, 36 pp

    work page 2019

  24. [24]

    Ho, The Brown measure of the sum of a self-adjoint element and an elliptic element, Electron

    C.-W. Ho, The Brown measure of the sum of a self-adjoint element and an elliptic element, Electron. J. Probab.27(2022), Paper No. 123, 32 pp

    work page 2022

  25. [25]

    Ho and P

    C.-W. Ho and P. Zhong, Brown measures of free circular and multiplicative Brownian motions with self-adjoint and unitary initial conditions,J. Eur. Math. Soc.25(2023), 2163–2227

    work page 2023

  26. [26]

    R. V. Kadison and J. R. Ringrose, Fundamentals of the theory of operator algebras. Vol. II, Grad. Stud. Math.,16American Mathematical Society, Providence, RI, 1997

    work page 1997

  27. [27]

    Introduction to smooth manifolds,

    J. Lee, “Introduction to smooth manifolds,” Second edition. Graduate Texts in Mathematics, 218, Springer, 2013

    work page 2013

  28. [28]

    Free probability and random matrices,

    J. Mingo and R. Speicher, “Free probability and random matrices,” Fields Institute Mono- graphs, 35, Springer; Fields Institute for Research in Mathematical Sciences, 2017

    work page 2017

  29. [29]

    Topology,

    J. R. Munkres, “Topology,” Second edition, Prentice Hall, 2000

    work page 2000

  30. [30]

    Twardowska, Wong-Zakai approximations for stochastic differential equations,Acta Appl

    K. Twardowska, Wong-Zakai approximations for stochastic differential equations,Acta Appl. Math.43(1996), 317–359

    work page 1996

  31. [31]

    Wong and M

    E. Wong and M. Zakai, On the convergence of ordinary integrals to stochastic integrals,Ann. Math. Statist.36(1965), 1560–1564

    work page 1965

  32. [32]

    Real analysis,

    J. Yeh, “Real analysis,” World Scientific Publishing Co., 2014

    work page 2014

  33. [33]

    Zhong, On the free convolution with a free multiplicative analogue of the normal distribu- tion,J

    P. Zhong, On the free convolution with a free multiplicative analogue of the normal distribu- tion,J. Theoret. Probab.28(2015), 1354–1379

    work page 2015

  34. [34]

    Zhong, Brown measure of the sum of an elliptic operator and a free random variable in a finite von Neumann algebra,Amer

    P. Zhong, Brown measure of the sum of an elliptic operator and a free random variable in a finite von Neumann algebra,Amer. J. Math., to appear. Department of Mathematics, University of Notre Dame, Notre Dame, IN 46556, USA Email address:bhall@nd.edu Institute of Mathematics, Academia Sinica, Taipei 10617, Taiwan Email address:chwho@gate.sinica.edu.tw

    work page