Pith. sign in

REVIEW 4 major objections 5 minor 38 references

Universality of Packing Dimension Estimates for Spectral Measures of Quasiperiodic Operators: Monotone Potentials

T0 review · 4 major / 5 minor · reviewed 2026-08-09 · 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 that the quantitative packing-dimension bound $2(1-L/\beta(\alpha))$ for spectral measures, previously known for the Almost Mathieu operator, holds for every quasiperiodic Schrödinger operator with a $\gamma$-monotone…

desk verdict A solid, well-written adaptation of the Jitomirskaya-Liu-Tcheremchantsev packing-dimension machinery to monotone potentials; the stress-test concern about d vs 1/d does not survive a close reading. read the letter →

arxiv 2502.03707 v2 pith:R2JDDZS2 submitted 2025-02-06 math.SP

Netanel Levi This is my paper · ORCID
classification math.SP MSC 47B3681Q1028A80
keywords quasiperiodicSchrödingeroperatormonotonepotentialpackingdimensionRényiLyapunovexponentsharparithmetictransitionspectralmeasureBoreltransform
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 proves that the quantitative fractal-dimension bounds previously found for the Almost Mathieu operator hold for every quasiperiodic Schrödinger operator with a $\gamma$-monotone potential, i.e. a 1-periodic potential satisfying $f(y)-f(x) \geq \gamma(y-x)$. The main result is: for any Borel set $A$ on which the Lyapunov exponent satisfies $L(E)<\beta(\alpha)$, with $L = \min_{E\in A} L(E)$, the upper packing dimension of the spectral measure restricted to $A$ is at most $2(1-L/\beta(\alpha))$; if $L(E)\geq\beta(\alpha)$ on $A$, that packing dimension is $0$. Here $\beta(\alpha)$ measures how well the frequency $\alpha$ is approximated by rationals. The same argument gives the Rényi-dimension bound $D^+_{\mu^x|A}(q) \leq (2\beta(\alpha)-2L)/(2\beta(\alpha)-L)$ for every $q\geq 3/2$ and every phase $x$. This matters because it shows the sharp arithmetic transition is accompanied by a universal, quantitative singularity of spectral measures, not just a qualitative change in spectral type.

What carries the argument

The argument rests on two interlocking estimates. First, Green's function decay for the finite-interval restrictions of $H$: results from the companion localization work ([20]) give exponential decay $e^{-(L(E)-o(1))|x-y|}$ for Green's functions away from 'resonant' points, with regularity scales tied to the continued-fraction denominators $q_n$ and to $\beta(\alpha)$. Second, this decay is converted by Proposition 3.10 into power-law lower bounds on the subordinacy quantity $\omega^{\pm}(L)=\max_{\theta}\|u_\theta\|^{\pm}_L \cdot \min_{\theta}\|u_\theta\|^{\pm}_L$, namely $\omega^{\pm}(L) \geq L^{1+(1/2)L(E)/(t_1\beta)-\varepsilon}$ (Lemma 4.2). Through the subordinacy theorem (Theorem 2.7) and a half-line Borel-transform lemma (Proposition 2.6), those lower bounds yield $\operatorname{Im} M(E+i\varepsilon) \geq \varepsilon^{-t}$ for $M$ the Borel transform of the spectral measure; Propositions 2.3 and 2.4 then convert the growth of $\operatorname{Im} M$ into the claimed upper bounds on $\gamma^+_\mu(E)$ and on $D^+_\mu(q)$. The named key quantity is the lower $\eta$-derivative $D^\eta_\mu(E)$, which is shown to be infinite for all $\eta$ above the threshold.

What would settle it

Take a $\gamma$-monotone potential with a coupling such that $L(E)<\beta(\alpha)$ for an energy $E$ near the spectrum, and compute the finite-volume Green's function $G_I(x,y)$ on intervals of length comparable to $q_n$ at scales where $q_{n+1}\approx e^{\beta(\alpha)q_n}$. If the estimate $|G_I(x,y)|\leq e^{-(L(E)-o(1))|x-y|}$ from Propositions 3.6 and 3.8 fails for some non-resonant point, while the Lyapunov exponent and $\beta(\alpha)$ are computable, then the key input to Theorem 3.1 is false. Alternatively, a direct check of Lemma 4.2's power-law growth of $\omega^+(L)$ would settle whether the almost-everywhere divergence $D^\eta_{\mu^x}(E)=\infty$ holds.

Watch

Extended reading notes

Core claim

The central claim is Theorem 3.1: for every $E$, writing $\Lambda(E)=\min\{L(E),\beta(\alpha)\}$, and for every $\eta>2(1-\Lambda(E)/\beta(\alpha))$, the lower $\eta$-derivative $D^\eta_{\mu^x}(E) = \liminf_{\varepsilon\to 0} \mu^x(E-\varepsilon,E+\varepsilon)/\varepsilon^\eta$ is infinite for $\mu^x$-almost every $E$, for every phase $x$. From this single almost-everywhere divergence statement the paper derives both the packing-dimension bound (Theorem 1.1) and the Rényi-dimension bound (Theorem 1.3), via general results linking such $\eta$-derivatives to upper local scaling exponents and to generalized Rényi sums. Thus the essence of the proof is to show that at almost every energy, the spectral measure is much more concentrated than any power $\varepsilon^\eta$ of the interval length, with the admissible $\eta$ controlled by the ratio $L(E)/\beta(\alpha)$. The dichotomy $L<\beta$ versus $L\geq\beta$ is exactly the sharp arithmetic transition studied for these operators; the paper's contribution is to show that it carries over to full fractal continuity properties for all monotone potentials.

Load-bearing premise

The proof relies on Green's-function decay estimates taken from the companion localization paper ([20]), but those estimates were originally proved for the regime $L(E)>\beta(\alpha)$, and Remark 3.13 simply asserts, without proof, that they remain valid when $L(E)\leq\beta(\alpha)$; if they degrade in that regime, the lower bounds on $\omega^{\pm}(L)$, the Borel-transform growth, and ultimately the packing-dimension bound collapse.

Editorial extensions

If this is right

  • For every $\gamma$-monotone potential, spectral measures on the set $\{E: L(E)<\beta(\alpha)\}$ have upper packing dimension at most $2(1-L/\beta(\alpha))$, with $L$ the minimum Lyapunov exponent on the set; this is the same quantitative bound as for the Almost Mathieu operator.
  • On sets where $L(E)\geq\beta(\alpha)$, the restricted spectral measures have upper packing dimension $0$, and when $L(E)>\beta(\alpha)$ the spectrum is purely atomic almost surely.
  • The Rényi-dimension estimate $D^+_{\mu^x|A}(q) \leq (2\beta(\alpha)-2L)/(2\beta(\alpha)-L)$ holds for all $q\geq 3/2$ and all phases, giving a uniform multifractal upper bound.
  • The meaningful nontrivial regime is $\beta(\alpha)/2 < L < \beta(\alpha)$; for $L\leq\beta(\alpha)/2$ the bound $2(1-L/\beta(\alpha))$ is at least $1$ and carries no information.

Reading between the lines

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

  • The same strategy should transfer to other families with a sharp arithmetic transition, for instance Type I operators or unitary analogues of the Almost Mathieu operator, assuming their Green's-function estimates also extend below the transition; if so, the packing-dimension bound is a genuinely universal feature of the transition, not an Almost-Mathieu-specific fact.
  • The threshold $\beta(\alpha)/2$ remains unaddressed: the theorem is vacuous there, so a plausible next question is whether the true upper packing dimension jumps to $1$ or stays below $1$ in that range for non-Almost-Mathieu monotone potentials.
  • The almost-everywhere formulation suggests a testable numerical signature: one could simulate the finite-volume transfer products for a monotone potential and extract the local scaling exponent $\gamma^+_\mu(E)$, comparing it to $2(1-L(E)/\beta(\alpha))$ over a fine energy grid.
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

4 major / 5 minor

Summary. The paper studies γ-monotone quasiperiodic Schrödinger operators on ℓ²(Z) and proves upper bounds for the upper packing dimension and upper Rényi dimensions of their spectral measures. The main result, Theorem 3.1, states that for every energy E with Λ(E)=min{L(E),β(α)}, the lower η-derivative D^η_{μ^x}(E) is infinite for μ^x-almost every E whenever η>2(1−Λ(E)/β(α)). From this, Theorem 1.1 derives the packing-dimension bound dim⁺_P(μ^x|A) ≤ 2(1−L/β(α)) when L(E)<β(α) on A, and Theorem 1.3 derives the Rényi-dimension bound D⁺_{μ^x|A}(q) ≤ (2β(α)−2L)/(2β(α)−L). The proof follows the strategy of [24] for the almost Mathieu operator, replacing the AMO-specific decay estimates by Green's-function regularity estimates from [20] for monotone potentials.

Significance. If the proof can be made fully rigorous, the result would be a substantial universality statement: the packing-dimension bound known for the almost Mathieu operator near the arithmetic transition would hold for the whole class of γ-monotone quasiperiodic potentials. The paper is clearly structured, with the central claim reduced to Theorem 3.1 and the auxiliary estimate Lemma 4.2, and it is a genuine extension rather than a repackaging of the AMO result. The explicit constants are parameter-free, and the abstract formulation in terms of Lyapunov exponent and β(α) gives clean, falsifiable predictions. On the other hand, the written proof currently contains a sign error in a key parameter and an unjustified inequality in the Green's-function estimate, and it relies on unpublished preprints for the main a priori estimates. These issues make the present version unsuitable for publication without substantial revision.

major comments (4)
  1. [Lemma 3.12, Eq. (3.10); Proposition 4.1, Eq. (4.4)] In the transition from (3.10) to (4.4), the proof concludes that d/2 ≤ e^{(1−t₁)β qₙ} from the lower bound d ≥ e^{−(β+o(1))qₙ + log|j|}. A lower bound on d cannot imply an upper bound on d/2 of this form; the inequality goes in the wrong direction. The claim d/2 ≤ e^{(1−t₁)β qₙ} may be recoverable from the upper bound d ≤ |j| ||qₙα|| ≤ |j|/q_{n+1} together with the restrictions on |j|, but this argument is not given. As written, the derivation of the prefactor in (4.4) is invalid, and that prefactor is essential for the iteration leading to Proposition 4.1.
  2. [Section 4, parameter t₁; Proposition 4.1; Lemma 4.2, Case 2] With t₁ defined as (β−L(E))/(β+σ), for L(E)<β(α) one computes L(E)−(1−t₁)β = σ(L(E)−β)/(β+σ) < 0. Thus the exponent in (4.2), −(1/2)(L(E)−(1−t₁)β−o(1))|k|, is positive for large |k|, so (4.2) is not a decay estimate and Proposition 4.1 is vacuous in the regime where it is used. Lemma 4.2, Case 2, needs L(E)−(1−t₁)β > 0 to obtain the positive exponential lower bound on the Wronskian solution, and without that both Lemma 4.2 and Theorem 3.1 fail. The definition of t₁ (or the sign in the exponent) must be corrected, and the correction must be consistent with the threshold computations in Appendix A.
  3. [Appendix A, proof of Lemma 4.3, formula for t₀] The displayed formula gives t₀ = (β−L(E))/(2β−L(E)+2σβ−ηβ−ηL(E)). As σ,η→0 this tends to (β−L(E))/(2β−L(E)), which is not always below the required threshold L(E)/(2β−L(E)); for example, when L(E)<β/2 the limit exceeds the threshold. Since Lemma 4.3 requires t₀ ∈ (t, L(E)/(2β−L(E))), the numerator likely should be L(E) rather than β−L(E), and t₁ must be chosen so that t₀ is strictly below the stated upper bound. This is a load-bearing step because Lemma 4.3 is the bridge from Lemma 4.2 to Theorem 3.1.
  4. [Section 3.2.2, Remark 3.13] The paper relies on Propositions 3.6, 3.8, and Lemma 3.12 from the unpublished preprint [20], and Remark 3.13 asserts without proof that these estimates remain valid when L(E) ≤ β(α). Since the main theorem is precisely concerned with that regime, the validity of the proof is contingent on an extension that is not demonstrated here. The author should either provide a proof of the extension or give a precise statement of where in [20] the L(E) ≤ β(α) case is covered, so that the referee and readers can verify the needed input.
minor comments (5)
  1. [Abstract and Section 1] The abstract says the monotone potential is 'as defined in [16]', while the introduction and Section 3.1 cite [20] for that definition; please make the reference consistent.
  2. [Section 3.1, γ-monotone definition] The definition of γ-monotone in (3.1) is written for 0 ≤ x < y < 1, but the potential f is allowed to take the value −∞; it would help to specify whether the condition is interpreted in the extended-real sense and how integrability of log(1+|f(x)|) is intended for unbounded values.
  3. [Proposition 4.1, around Eq. (4.3)] The notation r_{j±k} with k ∈ {1,2} is not defined; it should be spelled out as r_{j−2}, r_{j−1}, r_{j+1}, r_{j+2}, and the later definition M = max {r_k : k ∈ {−2,−1,0,1,2}} should indicate the dependence on p explicitly.
  4. [Lemma 4.2, Case 2 and Case 4] Several displayed inequalities in the proof of Lemma 4.2 are hard to parse: in Case 2, the chain 'e^{1/2(...)2q_n²q_{n+1}^{t₁}} ≥ 1/2 e^{L^{t₁/2}}' mixes powers of q_n and L in a way that needs clarification, and in Case 4 the step 'e^{(L(E)−ε)qₙ} L^{2−ε} = ((e^{qₙ})^{β+ε})^{(L(E)−ε)/(β+ε)}' should be rewritten with careful exponent arithmetic.
  5. [Appendices and references] There are typographical errors such as 'fucntions' in Section 2.3.3 and 'Jitomirsaya' in reference [16], and the proof of Lemma 4.3 contains phrases like 'letting η and σ go to 0' that should be replaced by a quantitative choice of small positive η,σ to guarantee the strict inequality t < t₀.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation: Theorem 1.1 and Theorem 1.3 are deduced from Theorem 3.1 via independent general lemmas; the cited external estimates are not self-referential, and the proof gaps identified by a skeptic are correctness issues, not circularity.

full rationale

The central claim is Theorem 3.1, which states that for every eta greater than 2(1 - Lambda(E)/beta(alpha)), the lower eta-derivative D^eta_{\mu^x}(E) is infinite for mu^x-almost every E. This theorem is the engine from which Theorem 1.1 and Theorem 1.3 are derived, using Proposition 2.2 and Proposition 2.4 respectively. Those propositions are quoted from [24] as general facts about packing dimension and Renyi dimension under pointwise lower bounds on the imaginary part of the Borel transform; they do not assume the packing-dimension conclusion being proved, so using them is legitimate external support, not a circular step. The quantities L(E), beta(alpha), Lambda(E), and mu^x are all defined independently in Sections 2.1, 2.3.4, and 3.1, and the target estimates are not inserted into those definitions. The author is not an author of the principal cited sources [20] and [24], so the self-citation-load-bearing and uniqueness-imported-from-authors patterns do not apply. The main weakness is Remark 3.13, where the paper asserts without proof that the Green's-function estimates from [20] remain valid when L(E) is at most beta(alpha), although [20] was developed for the localization regime L(E) greater than beta(alpha). This is an unproved extension and a genuine correctness risk, but it is not circular: the asserted estimate is an input that is stronger than the theorem being proved and is not equivalent to it by construction. Likewise, the skeptic's algebraic complaint about the factor d/2 in Lemma 3.12 versus the use of the reciprocal-type bound in Proposition 4.1 is a possible error in the proof of Proposition 4.1, not a case of a fitted parameter or a definition being renamed as a prediction. No equation in the paper is shown to reduce to its own input by definition, and no load-bearing claim is justified solely by a self-citation chain. Therefore the appropriate circularity score is minimal; the substantial concerns are about correctness and completeness of the external estimates, not circularity.

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

No parameters are fitted to data; the proof constants t1, t2, σ, τ, C are auxiliary and cancel from the final bounds. No new entities are introduced. The central claim rests on the cited decay estimates and standard spectral theory.

assumptions (4)
  • domain assumption γ-monotone potentials with ∫_0^1 log(1+|f(x)|)dx<∞ and β(α)<∞
    The theorems are stated only for this class; the decay estimates from [20] require monotonicity, integrability of log(1+|f|), and finite β(α).
  • standard math Standard continued fraction and Diophantine approximation facts about q_n and β(α)
    Section 2.1 uses the continued fraction approximation and limsup bounds on q_{n+1} in terms of q_n; these are classical.
  • standard math Spectral theory background: self-adjointness, cyclicity of {δ0,δ1}, Borel transform boundary behavior, subordinacy estimates
    Section 2.3 and Appendix A rely on these established results, including Propositions 2.3, 2.4, 2.6 and Theorem 2.7 from [24] and related references.
  • domain assumption The Green's function estimates of [20] (Propositions 3.6, 3.8, Lemma 3.12) extend verbatim to the regime L(E)≤β(α)
    Remark 3.13 states the extension without proof; Proposition 4.1 and Lemma 4.2 depend on it. This is the load-bearing assumption flagged in weakest_assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Universality of Packing Dimension Estimates for Spectral Measures of Quasiperiodic Operators: Monotone Potentials." pith.science (2026). https://pith.science/paper/R2JDDZS2

@misc{pith2026250203707,
  author       = {Pith},
  title        = {Pith review of: Universality of Packing Dimension Estimates for Spectral Measures of Quasiperiodic Operators: Monotone Potentials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R2JDDZS2}},
  note         = {Machine review of arXiv:2502.03707}
}
abstract

Let $H$ be a quasiperiodic Schr\"{o}dinger operator generated by a monotone potential, as defined in [16]. Following [20], we study the connection between the Lyapunov exponent $L\left(E\right)$, arithmetic properties of the frequency $\alpha$, and certain fractal-dimensional properties of the spectral measures of $H$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 37 canonical work pages

  1. [20]

    Sharp arithmetic localization for quasiperiodic operators with monotone potentials

    S. Jitomirskaya and I. Kachkovskiy, Sharp arithmetic localization for quasiperiodic operator s with monotone potentials., preprint. arXiv:2407.00703

    work page Pith review arXiv

  2. [24]

    Jitomirskaya, W

    S. Jitomirskaya, W. Liu and S. Tcheremchantsev, Lower bounds on concentration through Borel transform and quantitative singularity of spectral measur es near the arithmetic transition , preprint

    work page

  3. [1]

    Avila and S

    A. Avila and S. Jitomirskaya, The Ten Martini Problem , Ann. of Math. 170 (2009), 303–342

    work page 2009

  4. [2]

    Avila, J

    A. Avila, J. You and Q. Zhou, Sharp phase transitions for the almost Mathieu operator , Duke Math. J. 166 (2017), 2697–2718

    work page 2017

  5. [3]

    Ju. M. Berezanski ˘i, Expansions in Eigenfunctions of Self-Adjoint Operators , Transl. Math. Mono. 17, Amer. Math. Soc., Providence, RI, 1968

    work page 1968

  6. [4]

    Cedzich and J

    C. Cedzich and J. Fillman, Absence of bound states for quantum walks and CMV matrices vi a reflections, J. Spectr. Theory 14 (2024), 1513–1536

    work page 2024

  7. [5]

    Cedzich, J

    C. Cedzich, J. Fillman and D. Ong, Almost everything about the unitary almost Mathieu operato r, Commun. Math. Phys 403 (2023), 745–794

    work page 2023

  8. [6]

    C. D. Cutler, Measure disintegrations with respect to σ-stable monotone indices and the pointwise representation of packing dimension , Supp. Rend. Circ. Mat. Palermo 28 (1992), 319–339

    work page 1992

Show all 38 references
  1. [7]

    Damanik and J

    D. Damanik and J. Fillman, One-dimensional ergodic Schr¨ odinger operators–I. Gener al theory , Grad. Stud. Math. 221, American Mathematical Society, Providence, RI (2022)

    2022

  2. [8]

    Damanik and J

    D. Damanik and J. Fillman, One-dimensional ergodic Schr¨ odinger operators–II. Specific classes , Grad. Stud. Math. 221, American Mathematical Society, Providence, RI (2025)

    2025

  3. [9]

    Damanik, R

    D. Damanik, R. Killip and D. Lenz, Uniform Spectral Properties of One-Dimensional Quasicrys tals, III. α-continuity, Commun. Math. Phys 212 (2000), 191–204. 13

    2000

  4. [10]

    Damanik and S

    D. Damanik and S. Tcheremchantsev, Scaling estimates for solutions and dynamical lower bounds on wavepacket spreading , J. Anal. Math 97 (2005), 103–131

    2005

  5. [11]

    Del Rio, S

    R. Del Rio, S. Jitomirskaya, Y. Last and B. Simon, Operators with singular continuous spectrum, IV. Hausdorff dimensions, rank one perturbations, and local ization, J. Anal. Math 69 (1996), 153– 200

    1996

  6. [12]

    Ge and S

    L. Ge and S. Jitomirskaya, Hidden subcriticality, symplectic structure, and univers ality of sharp arithmetic spectral results for type I operators , preprint. arXiv:2407.08866

  7. [13]

    D. J. Gilbert and D. B. Pearson, On subordinacy and analysis of the spectrum of one-dimensio nal Schr¨ odinger operators, J. Math. Anal. Appl. 128 (1987), 30–56

    1987

  8. [14]

    Guarneri and H

    I. Guarneri and H. Schulz-Baldes, Lower bounds on wave packet propagation by packing dimensio ns of spectral measures, Math. Phys. Electron. J. 16 (1999), Paper 1

    1999

  9. [15]

    Jitomirskaya, Anderson localization for the almost Mathieu equation: a no nperturbative proof, Commun

    S. Jitomirskaya, Anderson localization for the almost Mathieu equation: a no nperturbative proof, Commun. Math. Phys. 165 (1994), 49–57

    1994

  10. [16]

    Jitomirsaya, Almost everything about the almost mathieu operator, II , Proc

    S. Jitomirsaya, Almost everything about the almost mathieu operator, II , Proc. of XI Int. Congress of Math. Physics, Int. Press, Somerville, Mass. (1995), 373–382

    1995

  11. [17]

    Jitomirskaya, Metal-insulator transition for the almost Mathieu operato r, Ann

    S. Jitomirskaya, Metal-insulator transition for the almost Mathieu operato r, Ann. of Math. 150 (1999), 1159–1175

    1999

  12. [18]

    Jitomirskaya and I

    S. Jitomirskaya and I. Kachkovskiy, All couplings localization for quasiperiodic operators wi th monotone potentials , J. Eur. Math. Soc. 21 (2019), 777–795

    2019

  13. [19]

    Jitomirskaya and I

    S. Jitomirskaya and I. Kachkovskiy, Sharp arithmetic delocalization for quasiperiodic operat ors with potentials of semi-bounded variation , preprint. arXiv:2408.16935

    arXiv

  14. [21]

    Jitomirskaya and Y

    S. Jitomirskaya and Y. Last, Power law subordinacy and singular spectra, I. Half-line op erators, Acta Math. 183 (1999), 171–189

    1999

  15. [22]

    Jitomirskaya and Y

    S. Jitomirskaya and Y. Last, Power law subordinacy and singular spectra, II. Line operat ors, Commun. Math. Phys. 211 (2000), 643–658

    2000

  16. [23]

    Jitomirskaya and W

    S. Jitomirskaya and W. Liu, Universal hierarchical structure of quasiperiodic eigenf unctions, Ann. of Math. 187 (2018), 721–776

    2018

  17. [25]

    Jitomirskaya and S

    S. Jitomirskaya and S. Zhang, Quantitative continuity of singular continuous spectral m easures and arithmetic criteria for quasiperiodic Schr¨ odinger op erators, J. Eur. Math. Soc. 24 (2022), 1723–1767

    2022

  18. [26]

    Kachkovskiy, Localization for quasiperiodic operators with unbounded m onotone potentials , J

    I. Kachkovskiy, Localization for quasiperiodic operators with unbounded m onotone potentials , J. Funct. Anal. 277 (2019), 3467–3490

    2019

  19. [27]

    Kerdboon and X

    J. Kerdboon and X. Zhu, Anderson localization for Schr¨ odinger operators with monotone potentials over circle homeomorphisms , J. Spectr. Theory 14 (2024), 1623–1646

    2024

  20. [28]

    Killip, A

    R. Killip, A. Kiselev and Y. Last, Dynamical upper bounds on wavepacket spreading , Amer. J. Math. 125 (2003), 1165–1198

    2003

  21. [29]

    Khan and D

    S. Khan and D. B. Pearson, Subordinacy and spectral theory for infinite matrices , Helv. Phys. Acta 65 (1992), 505–527. 14

    1992

  22. [30]

    Last and B

    Y. Last and B. Simon, Eigenfunctions, transfer matrices, and absolutely contin uous spectrum of one-dimensional Schr¨ odinger operators, Invent. Math. 135 (1999), 329–367

    1999

  23. [31]

    Liu and X

    W. Liu and X. Yuan, Anderson localization for the almost Mathieu operator in th e exponential regime, J. Spectr. Theory 5 (2015), 89–112

    2015

  24. [32]

    Liu and X

    W. Liu and X. Yuan, Anderson localization for the completely resonant phases , J. Funct. Anal. 268 (2015), 732–747

    2015

  25. [33]

    C. A. Rogers and S. J. Taylor, The analysis of additive set functions in Euclidean space , Acta Math. 101 (1959), 273–302

    1959

  26. [34]

    Proc. Mathematical Quantum Theory, II: Schr¨ odinger Operators

    B. Simon, Spectral analysis of rank one perturbations and applicatio ns, in “Proc. Mathematical Quantum Theory, II: Schr¨ odinger Operators” (Vancouver, Canada, 1993), CRM Proceedings and Lecture Notes, 8, American Mathematical Society, Providence, R I, 1995, 109–149

    1993

  27. [35]

    Yang, Anderson localization for the unitary almost Mathieu opera tor, Nonlinearity 37 (2024), 085010

    F. Yang, Anderson localization for the unitary almost Mathieu opera tor, Nonlinearity 37 (2024), 085010. A Proof of Lemma 4.3 We prove Lemma 4.3 following the lines of [24, Lemma 9.2]. Proof of Lemma 4.3. We will prove that there exists θ ∈ [0,π ) such that Imm+ θ (E +iε) ≥ε−t...

    2024

  28. [36]

    S is the union of qn disjoint intervals of length 2 9qn , each such interval contains exactly two points from S

  29. [37]

    The distance between two points in the same interval is |j| dist (qnα, Z)

  30. [38]

    Let Λ = σ (H2qn−1)

    The 1 9qn -neighborhoods of these intervals are disjoint. Let Λ = σ (H2qn−1). Since |Λ | ≤ 2qn − 1, there exists at least one interval K such that |K ∩ Λ | ≤ 1. Note that this interval K contains two points x0 = p0α,x 1 = p1α which satisfy the assumptions of Threom B.1 with l ...

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.