arxiv: 2605.00478 · v1 · submitted 2026-05-01 · 🧮 math-ph · math.MP

Recognition: unknown

Strong-disorder expansion of the root-averaged density of states for the Anderson model on the Bethe lattice

Masahiro Kaminaga

Authors on Pith no claims yet

Pith reviewed 2026-05-09 18:48 UTC · model grok-4.3

classification 🧮 math-ph math.MP

keywords Anderson modelBethe latticedensity of statesstrong disorderrandom walk expansionStieltjes transformroot-averaged spectral measureanalytic continuation

0 comments

The pith

The root-averaged density of states for the Anderson model on the Bethe lattice is absolutely continuous with a real-analytic strong-disorder expansion inside a scaled energy window.

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

The paper proves that when the single-site random potential has compact support and locally analytic density, the averaged diagonal resolvent for the Anderson Hamiltonian on the infinite Bethe lattice admits a holomorphic extension to a neighborhood of a given energy interval once the disorder strength exceeds a threshold. Stieltjes inversion then yields that the root-averaged spectral measure is absolutely continuous on the scaled window λI, with a density that is real analytic in energy and admits a power-series expansion in 1/λ. The leading term equals the single-site density, all odd powers vanish identically, and every higher coefficient is a finite sum over occupation profiles of short closed walks on the tree. For the uniform distribution the first correction is computed explicitly.

Core claim

The scaled averaged diagonal resolvent has a holomorphic continuation to a complex neighborhood of I for all sufficiently large λ. By the Stieltjes inversion formula, the root-averaged density of states measure is absolutely continuous on the scaled energy window λI, and its density is real analytic and has a finite-order strong-disorder expansion there. In the scaled form E=λξ, the leading coefficient is the local density of the single-site distribution. All odd coefficients vanish, and the higher coefficients are finite sums determined by occupation profiles of short closed walks on the tree.

What carries the argument

Random-walk expansion of the resolvent on the tree combined with holomorphic continuation of the single-site Stieltjes transforms.

If this is right

The density of states admits an explicit asymptotic series in inverse disorder strength whose coefficients are computable from finite walk data.
All odd-order terms in the expansion are identically zero.
For any fixed energy inside the scaled window the density converges to the single-site density as disorder tends to infinity.
The first nonzero correction for the uniform distribution is given by a concrete sum over two-step closed walks.

Where Pith is reading between the lines

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

The same analyticity and expansion structure should hold for the Anderson model on any regular tree of degree greater than two.
The absence of odd powers suggests a hidden even symmetry in the moments of the resolvent under strong disorder.
The walk-profile representation may extend to compute higher moments or the integrated density of states without additional analytic effort.

Load-bearing premise

The single-site distribution has compact support and possesses a locally analytic density on an interval containing the energy window of interest.

What would settle it

A numerical diagonalization or recursion computation on large finite Bethe lattices showing that the root-averaged density of states fails to be real analytic at some point inside λI for arbitrarily large λ.

Figures

Figures reproduced from arXiv: 2605.00478 by Masahiro Kaminaga.

**Figure 1.** Figure 1: Schematic illustration of I, I ♯ , and the lower boundary arc η. The contour used in the proof is I ♯ ∪ (−η). Lemma 4.1. For each k = 1, 2, . . . , the function sk defined on C+ by (1) extends holomorphically to Ωδ(I). More precisely, for ζ ∈ Ωδ(I), sk(ζ) = Z R\I ♯ dµ(t) (t − ζ) k + Z η ρ(w) dw (w − ζ) k . (18) Moreover, there exists a constant Cδ > 0 such that |sk(ζ)| ≤ Cδ(δ0 − δ) −k , ζ ∈ Ωδ(I), k = 1, … view at source ↗

read the original abstract

We study the root-averaged density of states for the Anderson model on the Bethe lattice in the strong-disorder regime. Here the density of states means the root-averaged spectral measure, not a finite-volume eigenvalue counting limit. We assume that the single-site distribution has compact support and has a locally analytic density on an interval $I^\sharp$ containing a given interval $I$. Combining the random-walk expansion on the tree with a complex-analytic argument for the single-site Stieltjes transforms, we prove that the scaled averaged diagonal resolvent has a holomorphic continuation to a complex neighborhood of $I$ for all sufficiently large $\lambda$. By the Stieltjes inversion formula, the root-averaged density of states measure is absolutely continuous on the scaled energy window $\lambda I$, and its density is real analytic and has a finite-order strong-disorder expansion there. In the scaled form $E=\lambda\xi$, the leading coefficient is the local density of the single-site distribution. All odd coefficients vanish, and the higher coefficients are finite sums determined by occupation profiles of short closed walks on the tree. For the uniform single-site distribution, we compute the first nonzero correction term explicitly.

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.

Desk Editor's Note private letter to a colleague

The paper proves a clean strong-disorder expansion for the root-averaged density of states on the Bethe lattice, with odd coefficients vanishing by the tree structure.

read the letter

The main result is a rigorous finite-order expansion in powers of 1/λ for the root-averaged density of states of the Anderson model on the Bethe lattice in the strong-disorder regime. After scaling the energy as E = λξ, the density is real analytic on the interval and its coefficients are determined by occupation numbers of short closed walks on the tree. All odd powers drop out, the leading term is just the single-site density, and the first correction is computed explicitly for the uniform distribution. This is obtained by combining the random-walk expansion of the resolvent with holomorphic continuation of the single-site Stieltjes transforms to a fixed complex neighborhood, followed by Stieltjes inversion. The argument is direct and uses only standard properties of resolvents and random walks on trees, with no circular definitions or hidden λ-dependence in the neighborhood radius. The assumptions (compact support plus local analyticity of the single-site density on a slightly larger interval) are stated up front and are necessary for the analytic continuation step. That analyticity condition is the clearest limitation; it excludes merely smooth or less regular distributions, but the paper does not claim to cover those cases. Within the stated class the estimates appear tight and the vanishing of odd terms follows naturally from the parity properties of the walk contributions on the tree. Readers already familiar with random-walk expansions for tree operators will see the new piece as the controlled analytic continuation and the resulting explicit coefficient formulas. The work is internally consistent and organizes the asymptotics in a regime of interest for disordered systems on trees. It deserves a serious referee because the combination of tools produces a new explicit result with verifiable structure.

Referee Report

0 major / 3 minor

Summary. The paper claims that for the Anderson model on the Bethe lattice, in the strong-disorder regime, the root-averaged density of states measure on the scaled interval λI is absolutely continuous with a real-analytic density admitting a finite-order expansion in powers of 1/λ. Under the assumptions that the single-site distribution has compact support and a locally analytic density on an interval I♯ ⊃ I, the scaled root-averaged diagonal resolvent admits a holomorphic continuation to a fixed complex neighborhood of I for large λ. This is established via random-walk expansion on the tree combined with analytic continuation of single-site Stieltjes transforms. The leading coefficient in the expansion (E = λξ) is the local single-site density; all odd coefficients vanish; higher coefficients are finite sums determined by occupation profiles of short closed walks. An explicit first nonzero correction is computed for the uniform distribution.

Significance. If the result holds, the manuscript supplies a rigorous, non-perturbative justification for the strong-disorder expansion of the root-averaged DOS on the Bethe lattice, a model central to Anderson localization studies. The explicit walk-profile representation of the coefficients yields falsifiable, computable predictions, and the combination of random-walk expansions with Stieltjes inversion and holomorphic continuation constitutes a technically clean application of established tools. This advances analytic control of spectral measures beyond finite-volume or numerical approaches and may serve as a template for related tree-like or high-dimensional disordered systems.

minor comments (3)

[Introduction] §1 (Introduction): the precise definition of the root-averaged diagonal resolvent G_λ(ξ) and its relation to the spectral measure should be stated explicitly before the main theorem, to avoid any ambiguity with the usual finite-volume DOS.
[§3] §3 (Random-walk expansion): the error bound for truncating the walk expansion at finite length is stated but the dependence of the remainder on the neighborhood radius is not quantified; a short remark on uniformity in the complex neighborhood would strengthen the analytic-continuation step.
[§4] The explicit first-order term for the uniform distribution (presumably in §4 or an appendix) is a valuable addition; including the numerical value of the coefficient alongside the walk-profile expression would aid immediate verification.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the careful and positive assessment of our manuscript. The report accurately summarizes our main results on the strong-disorder expansion of the root-averaged density of states for the Anderson model on the Bethe lattice. We note the recommendation for minor revision; however, the report contains no specific major comments requiring detailed rebuttal. We will incorporate any minor editorial improvements in the revised version.

Circularity Check

0 steps flagged

No significant circularity; self-contained analytic proof

full rationale

The derivation proceeds directly from the stated assumptions (compact support and local analyticity of the single-site density on I♯ ⊃ I) via random-walk expansion of the resolvent on the Bethe lattice and holomorphic continuation of single-site Stieltjes transforms to a fixed complex neighborhood of the scaled interval for large λ. The Stieltjes inversion formula then yields absolute continuity with real-analytic density whose strong-disorder expansion coefficients are explicitly finite sums over occupation profiles of short closed walks; the leading term is the single-site density and odd terms vanish by symmetry. No step defines a quantity in terms of the claimed result, renames a fitted parameter as a prediction, or reduces the central claim to a self-citation chain or ansatz imported from the authors' prior work. The argument is self-contained against external benchmarks of resolvent theory and tree random walks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on one domain assumption about the single-site distribution and on standard mathematical facts about Stieltjes transforms and random walks on trees; no free parameters are introduced or fitted.

axioms (1)

domain assumption The single-site distribution has compact support and has a locally analytic density on an interval I♯ containing a given interval I.
Explicitly stated in the abstract as the hypothesis required for the holomorphic continuation and the resulting analyticity of the density.

pith-pipeline@v0.9.0 · 5507 in / 1424 out tokens · 47255 ms · 2026-05-09T18:48:39.590044+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

14 extracted references · 11 canonical work pages

[1]

Acosta and A

V. Acosta and A. Klein, Analyticity of the density of stat es in the Anderson model on the Bethe lattice, J. Stat. Phys. , vol. 69, nos. 1-2, pp. 277-305, 1992. DOI: https://doi.org/ 10.1007/BF01053794

work page doi:10.1007/bf01053794 1992
[2]

Aizenman, R

M. Aizenman, R. Sims, and S. Warzel, Stability of the abso lutely continuous spectrum of random Schr¨ odinger operators on tree graphs, Probab. Theory Related Fields , vol. 136, no. 3, pp. 363-394, 2006. DOI: https://doi.org/10.1007/s00440-005-0486-8

work page doi:10.1007/s00440-005-0486-8 2006
[3]

Aizenman and S

M. Aizenman and S. Warzel, Resonant delocalization for r andom Schr¨ odinger operators on tree graphs, J. Eur. Math. Soc. , vol. 15, no. 4, pp. 1167-1222, 2013. DOI: https: //doi.org/10.4171/JEMS/389

work page doi:10.4171/jems/389 2013
[4]

J. V. Bellissard and P. D. Hislop, Smoothness of correlat ions in the Anderson model at strong disorder, Ann. Henri Poincar´ e, vol. 8, pp. 1-26, 2007. DOI: https://doi.org/10. 1007/s00023-006-0297-1

2007
[5]

Statistical theory of superlattices,

H. A. Bethe, “Statistical theory of superlattices,” Proc. Roy. Soc. London Ser. A , vol. 150, no. 871, pp. 552-575, 1935. DOI: https://doi.org/10.1098/rspa.1935.0122

work page doi:10.1098/rspa.1935.0122 1935
[6]

Bovier, M

A. Bovier, M. Campanino, A. Klein, and J. F. Perez, Smooth ness of the density of states in the Anderson model at high disorder, Commun. Math. Phys. , vol. 114, no. 3, pp. 439-461,
[7]

DOI: https://doi.org/10.1007/BF01242138

work page doi:10.1007/bf01242138
[8]

Constantinescu, J

F. Constantinescu, J. Fr¨ ohlich, and T. Spencer, Analyt icity of the density of states and replica method for random Schr¨ odinger operators on a latti ce, J. Stat. Phys. , vol. 34, nos. 3-4, pp. 571-596, 1984. DOI: https://doi.org/10.1007/BF01018559. 15

work page doi:10.1007/bf01018559 1984
[9]

H. L. Cycon, R. G. Froese, W. Kirsch, and B. Simon, Schr¨ odinger Operators with Appli- cation to Quantum Mechanics and Global Geometry , Springer-Verlag, Berlin, 1987. DOI: https://doi.org/10.1007/978-3-540-77522-5

work page doi:10.1007/978-3-540-77522-5 1987
[10]

D. R. Dolai and M. Krishna, Smoothness of integrated dens ity of states of the Anderson model on Bethe lattice in high disorder, J. Stat. Phys. , vol. 190, no. 5, article 103, 2023. DOI: https://doi.org/10.1007/s10955-023-03114-6

work page doi:10.1007/s10955-023-03114-6 2023
[11]

Kaminaga, M

M. Kaminaga, M. Krishna, and S. Nakamura, A note on the an alyticity of density of states, J. Stat. Phys. , vol. 149, no. 3, pp. 496-504, 2012. DOI: https://doi.org/10.1007/ s10955-012-0603-x

2012
[12]

Kirsch and B

W. Kirsch and B. Metzger, The integrated density of stat es for random Schr¨ odinger op- erators, in Spectral Theory and Mathematical Physics: A Festschrift in H onor of Barry Simon ’s 60th Birthday, Proc. Sympos. Pure Math., vol. 76, Part 2, pp. 649-696, Amer i- can Mathematical Society, Providence, RI, 2007. DOI: https://doi.org/10.1090/pspum/ 076.2/2307751

work page doi:10.1090/pspum/ 2007
[13]

Klein, Absolutely continuous spectrum in the Anders on model on the Bethe lattice, Math

A. Klein, Absolutely continuous spectrum in the Anders on model on the Bethe lattice, Math. Res. Lett. , vol. 1, no. 4, pp. 399-407, 1994. DOI: https://doi.org/10.4310/MRL. 1994.v1.n4.a1

work page doi:10.4310/mrl 1994
[14]

Klein, Extended states in the Anderson model on the Be the lattice, Adv

A. Klein, Extended states in the Anderson model on the Be the lattice, Adv. Math., vol. 133, no. 1, pp. 163-184, 1998. DOI: https://doi.org/10.1006/aima.1997.1688. Statements and Declarations Competing Interests The author declares that he has no competing interests. Data availability No datasets were generated or analyzed during the current st udy. Fundi...

work page doi:10.1006/aima.1997.1688 1998