pith. machine review for the scientific record. sign in

arxiv: 2604.20215 · v1 · submitted 2026-04-22 · 🧮 math.PR · math-ph· math.MP· math.SP

Recognition: unknown

Edge Universality for Inhomogeneous Random Matrices II: Markov Chain Comparison and Critical Statistics

Dang-Zheng Liu , Guangyi Zou
Authors on Pith no claims yet

Pith reviewed 2026-05-09 22:46 UTC · model grok-4.3

classification 🧮 math.PR math-phmath.MPmath.SP
keywords edge universalityinhomogeneous random matricesMarkov chain comparisonrandom band matricescritical statisticsvariance profilespectral edgeuniversality in random matrices
0
0 comments X

The pith

Two inhomogeneous random matrices exhibit the same universal edge statistics if their variance-profile Markov chains are comparable.

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

The paper develops short-to-long comparison conditions that extend edge universality results to subcritical and critical sparsity regimes for inhomogeneous random matrices. It proves that comparability of the variance-profile Markov chains is sufficient for the matrices to share identical universal edge statistics, independent of entry details. Applications to random band matrices, the Wegner orbital model, and Hankel-profile matrices demonstrate both universal and non-universal edge behaviors. A reader cares because this provides a general method to determine edge statistics by checking chain comparability rather than computing each model separately.

Core claim

We prove that two inhomogeneous random matrices exhibit the same universal edge statistics whenever their variance-profile Markov chains are comparable, regardless of the fine details of the matrix entries. This is achieved by developing new short-to-long comparison conditions that extend the mixing condition from the first paper in the series to the subcritical and critical sparsity regimes. The theorem is used to derive the spectral edge statistics for random band matrices, the Wegner orbital model, and Hankel-profile random matrices, uncovering a rich landscape of both universal and non-universal phenomena shaped by geometric structure, spike patterns, and domains of stable attraction.

What carries the argument

Short-to-long comparison conditions on variance-profile Markov chains that establish equivalence of edge statistics between models.

If this is right

  • The edge statistics of random band matrices follow from comparison to models with known universality.
  • The Wegner orbital model exhibits universal edge statistics under the comparability condition.
  • Hankel-profile random matrices can display non-universal critical statistics depending on their profile.
  • Both universal and non-universal phenomena arise from the geometric structure and domains of stable attraction of the Markov chains.

Where Pith is reading between the lines

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

  • This suggests that the edge behavior is governed by the long-term mixing properties of the variance profile rather than local entry distributions.
  • The method opens the way to analyze edge statistics in other structured random matrices by constructing appropriate Markov chain comparisons.
  • Testable extensions include verifying the conditions numerically for additional models like sparse Wigner matrices or block models.

Load-bearing premise

The short-to-long comparison conditions hold for the variance-profile Markov chains of the two inhomogeneous random matrices being compared.

What would settle it

A concrete counterexample would be two random matrices with comparable variance-profile Markov chains but different limiting edge distributions for their largest eigenvalues.

Figures

Figures reproduced from arXiv: 2604.20215 by Dang-Zheng Liu, Guangyi Zou.

Figure 1
Figure 1. Figure 1: Example of one possible gluing in Her￾mitian case of E[Tr(X6 )Tr(X4 )Tr(X8 )]. The green vertices are marked vertices. V1 V2 V1 V2 [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: Step 1: Remove all edges in F c , leaving a spanning forest F where each tree has exactly one “open” root vertex. Step 2: Sum over labels on each tree (collapsing them), so only the open edges and open vertices remain. = (1 + a n )r |Vb| (bn) |Eint|−|Vint| n |Eb|+|Vint| (|Eb| + |Vint|)!. (2.12) Here we sum over the evaluation of η along the forest in the second equality and use (2.5) in the third inequalit… view at source ↗
Figure 4
Figure 4. Figure 4: Comparison of the largest eigenvalue distributions for [PITH_FULL_IMAGE:figures/full_fig_p043_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Localization–delocalization transition of the eigenvector [PITH_FULL_IMAGE:figures/full_fig_p044_5.png] view at source ↗
read the original abstract

The first paper in this series introduced a \emph{short-to-long mixing} condition that captures mean-field GOE/GUE edge universality in the supercritical sparsity regime, for symmetric/Hermitian random matrices with independent entries and a Markov variance profile. This condition reduces the universality problem to the mixing properties of the underlying Markov chains. In this paper, we develop new \emph{short-to-long comparison} conditions that extend the analysis to the subcritical and critical sparsity regimes. Specifically, we prove that two inhomogeneous random matrices exhibit the same universal edge statistics whenever their variance-profile Markov chains are comparable, regardless of the fine details of the matrix entries. To illustrate the power of our Markov chain comparison theorem, we derive the spectral edge statistics for several prototypical models: random band matrices, the Wegner orbital model, and Hankel-profile random matrices. These comparisons uncover a rich landscape of both universal and non-universal phenomena -- shaped by geometric structure, spike patterns, and domains of stable attraction -- features that lie fundamentally beyond the reach of classical random matrix theory.

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

3 major / 2 minor

Summary. The paper develops short-to-long comparison conditions for the Markov chains associated with variance profiles of inhomogeneous random matrices. It proves that two such matrices share the same universal edge statistics whenever their variance-profile Markov chains are comparable, extending the short-to-long mixing framework from part I to subcritical and critical sparsity regimes. Applications to random band matrices, the Wegner orbital model, and Hankel-profile matrices are used to illustrate both universal and non-universal edge behaviors determined by geometric structure, spike patterns, and domains of stable attraction.

Significance. If the central comparison theorem holds with the stated error controls, the work supplies a flexible reduction of edge universality to Markov chain mixing properties that applies beyond mean-field regimes. This framework reveals phenomena inaccessible to classical RMT and provides a systematic way to classify critical statistics for structured sparse matrices. The explicit applications to prototypical models constitute a concrete strength.

major comments (3)
  1. [§3, Theorem 3.2] §3, Theorem 3.2 (comparison theorem): the short-to-long comparison condition is formulated in terms of total-variation distances between the chains, but the quantitative bound on the discrepancy in the edge eigenvalue distribution (the O(·) term in the statement) is not derived explicitly for the critical sparsity regime; this control is load-bearing for the claim that the statistics coincide regardless of entry details.
  2. [§5.1] §5.1, application to random band matrices: the Markov chain for the band variance profile is asserted to be comparable to the GOE chain, yet the mixing-time estimate used to verify the comparison condition is only sketched and does not include the dependence on the band width parameter in the subcritical regime.
  3. [§5.3] §5.3, Hankel-profile matrices: the claimed non-universal behavior arising from the domain of stable attraction is supported only by a heuristic argument; no rigorous perturbation or moment comparison is supplied to separate it from the universal case, which is central to the paper’s assertion of a “rich landscape.”
minor comments (2)
  1. [§2] The notation for the variance profile matrix V and the associated Markov transition kernel P is introduced in §2 but reused with slight variations in the applications; a single consolidated definition would improve readability.
  2. [Introduction] Several statements in the introduction refer to “the first paper in this series” without a precise citation label; add the arXiv number or section reference for part I.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thorough review and valuable suggestions. We address each major comment below and will incorporate clarifications and expansions in the revised version.

read point-by-point responses

  1. Referee: [§3, Theorem 3.2] §3, Theorem 3.2 (comparison theorem): the short-to-long comparison condition is formulated in terms of total-variation distances between the chains, but the quantitative bound on the discrepancy in the edge eigenvalue distribution (the O(·) term in the statement) is not derived explicitly for the critical sparsity regime; this control is load-bearing for the claim that the statistics coincide regardless of entry details.

    Authors: In the proof of Theorem 3.2 the discrepancy is controlled by combining the short-to-long comparison hypothesis with the edge universality established in Part I; the resulting O(·) term is bounded by the total-variation distance and remains valid under the critical-regime assumptions on the variance profiles. We agree that the explicit dependence should be displayed more clearly. In the revision we will add a short lemma immediately after the theorem that isolates and computes this error term for the critical case. revision: yes

  2. Referee: [§5.1] §5.1, application to random band matrices: the Markov chain for the band variance profile is asserted to be comparable to the GOE chain, yet the mixing-time estimate used to verify the comparison condition is only sketched and does not include the dependence on the band width parameter in the subcritical regime.

    Authors: The mixing-time estimate in §5.1 is presented in condensed form. The band-matrix variance profile corresponds to a random walk on the path graph with step size at most the band width w; standard conductance bounds then yield a mixing time of order (N/w)^2, which satisfies the short-to-long condition throughout the subcritical range. We will expand the sketch into a self-contained paragraph that records the explicit w-dependence and the resulting total-variation bound. revision: yes

  3. Referee: [§5.3] §5.3, Hankel-profile matrices: the claimed non-universal behavior arising from the domain of stable attraction is supported only by a heuristic argument; no rigorous perturbation or moment comparison is supplied to separate it from the universal case, which is central to the paper’s assertion of a “rich landscape.”

    Authors: Non-universality for the Hankel profile follows directly from the fact that its associated Markov chain fails the short-to-long comparability condition with the GOE chain, because the transition kernel produces a different domain of attraction. The calculation in §5.3 is heuristic in that it relies on an explicit moment expansion of the variance profile; the main comparison theorem already guarantees that the edge statistics cannot coincide with the universal case. To strengthen the presentation we will add a rigorous moment comparison in an appendix that computes the first few edge moments for the Hankel ensemble and contrasts them with the GOE moments. revision: yes

Circularity Check

0 steps flagged

Minor self-reference to series predecessor; core comparison theorem and applications remain independent

full rationale

The derivation introduces new short-to-long comparison conditions that extend the mixing framework from the prior paper in the series to subcritical/critical regimes. The central result—that comparable variance-profile Markov chains imply identical edge statistics—is stated and applied directly to models such as random band matrices, the Wegner orbital model, and Hankel-profile matrices without reducing any prediction or uniqueness claim to a fitted input or self-citation chain. The single reference to the first paper supplies background context for the mixing condition but does not bear the load of the new comparison theorem or the exhibited universal/non-universal phenomena. No self-definitional, fitted-prediction, or ansatz-smuggling reductions appear in the stated claims or proof outline.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Central claim rests on the newly introduced comparison conditions for Markov chains and the independent-entries assumption with given variance profiles; no free parameters or invented entities are introduced.

axioms (1)
  • domain assumption Independent entries with prescribed variance profile whose associated Markov chain satisfies the short-to-long comparison conditions.
    Stated as the basis for reducing universality to chain comparability, extending the mixing condition of the first paper.

pith-pipeline@v0.9.0 · 5491 in / 1132 out tokens · 15980 ms · 2026-05-09T22:46:58.847482+00:00 · methodology

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

64 extracted references · 17 canonical work pages

  1. [1]

    The O xford handbook of random matrix theory

    Gernot Akemann, Jinho Baik, and Philippe Di Francesco, editors. The O xford handbook of random matrix theory . Oxford University Press, Oxford, 2011

    2011

  2. [2]

    Bandeira, March T

    Afonso S. Bandeira, March T. Boedihardjo, and Ramon van Handel. Matrix concentration inequalities and free probability. Invent. Math. , 234(1):419--487, 2023

    2023

  3. [3]

    Bandeira, Giorgio Cipolloni, Dominik Schr¨ oder, and Ramon van Handel,Matrix concentration inequalities and free probability II

    Afonso S Bandeira, Giorgio Cipolloni, Dominik Schr \"o der, and Ramon van Handel. Matrix concentration inequalities and free probability II . two-sided bounds and applications. arXiv preprint arXiv:2406.11453 , 2024

    work page arXiv 2024

  4. [4]

    Largest eigenvalues and eigenvectors of band or sparse random matrices

    Florent Benaych-Georges and Sandrine P\'ech\'e. Largest eigenvalues and eigenvectors of band or sparse random matrices. Electron. Commun. Probab. , 19:no. 4, 9, 2014

    2014

  5. [5]

    Bulk eigenvalue statistics for random regular graphs

    Roland Bauerschmidt, Jiaoyang Huang, Antti Knowles, and Horng-Tzer Yau. Bulk eigenvalue statistics for random regular graphs. Ann. Probab. , 45(6A):3626--3663, 2017

    2017

  6. [6]

    Random band matrices

    Paul Bourgade. Random band matrices. In Proceedings of the I nternational C ongress of M athematicians--- R io de J aneiro 2018. V ol. IV . I nvited lectures , pages 2759--2784. World Sci. Publ., Hackensack, NJ, 2018

    2018

  7. [7]

    Topics in mathematics of data science

    Afonso S Bandeira, Amit Singer, and Thomas Strohmer. Topics in mathematics of data science. Preprint available online: https://people. math. ethz. ch/ abandeira//BandeiraSingerStrohmer-MDS-draft. pdf , 2025

    2025

  8. [8]

    Extremal random matrices with independent entries and matrix superconcentration inequalities

    Tatiana Brailovskaya and Ramon van Handel. Extremal random matrices with independent entries and matrix superconcentration inequalities. arXiv preprint arXiv:2401.06284 , 2024

    work page arXiv 2024

  9. [9]

    Universality and sharp matrix concentration inequalities

    Tatiana Brailovskaya and Ramon van Handel. Universality and sharp matrix concentration inequalities. Geom. Funct. Anal. , 34(6):1734--1838, 2024

    2024

  10. [10]

    Random band matrices in the delocalized phase I : Quantum unique ergodicity and universality

    Paul Bourgade, Horng Tzer Yau, and Jun Yin. Random band matrices in the delocalized phase I : Quantum unique ergodicity and universality. Communications on Pure and Applied Mathematics , 73(7):1526--1596, 2020

    2020

  11. [11]

    Eigenstate thermalization hypothesis

    Joshua M Deutsch. Eigenstate thermalization hypothesis. Reports on Progress in Physics , 81(8):082001, 2018

    2018

  12. [12]

    From quantum chaos and eigenstate thermalization to statistical mechanics and thermodynamics

    Luca D'Alessio, Yariv Kafri, Anatoli Polkovnikov, and Marcos Rigol. From quantum chaos and eigenstate thermalization to statistical mechanics and thermodynamics. Advances in Physics , 65(3):239--362, 2016

    2016

  13. [13]

    Local operator theory, random matrices and banach spaces

    Kenneth R Davidson and Stanislaw J Szarek. Local operator theory, random matrices and banach spaces. In Handbook of the geometry of Banach spaces , volume 1, pages 317--366. Elsevier, 2001

    2001

  14. [14]

    Probability---theory and examples , volume 49 of Cambridge Series in Statistical and Probabilistic Mathematics

    Rick Durrett. Probability---theory and examples , volume 49 of Cambridge Series in Statistical and Probabilistic Mathematics . Cambridge University Press, Cambridge, fifth edition, 2019

    2019

  15. [15]

    D. J. Daley and D. Vere-Jones. An introduction to the theory of point processes. V ol. I . Probability and its Applications (New York). Springer-Verlag, New York, second edition, 2003. Elementary theory and methods

    2003

  16. [16]

    Dubova, F

    Sofiia Dubova, Fan Yang, Horng-Tzer Yau, and Jun Yin. Delocalization of non-mean-field random matrices in dimensions d 3 . arXiv preprint arXiv:2507.20274 , 2025

    work page arXiv 2025

  17. [17]

    Dubova, F

    Sofiia Dubova, Kevin Yang, Horng-Tzer Yau, and Jun Yin. Delocalization of two-dimensional random band matrices. arXiv preprint arXiv:2503.07606 , 2025

    work page arXiv 2025

  18. [18]

    Quantum diffusion and delocalization for band matrices with general distribution

    L\'aszl\'o Erd\"os and Antti Knowles. Quantum diffusion and delocalization for band matrices with general distribution. Ann. Henri Poincar\'e , 12(7):1227--1319, 2011

    2011

  19. [19]

    Quantum diffusion and eigenfunction delocalization in a random band matrix model

    L\'aszl\'o Erd\"os and Antti Knowles. Quantum diffusion and eigenfunction delocalization in a random band matrix model. Comm. Math. Phys. , 303(2):509--554, 2011

    2011

  20. [20]

    Delocalization and diffusion profile for random band matrices

    L\'aszl\'o Erd o s, Antti Knowles, Horng-Tzer Yau, and Jun Yin. Delocalization and diffusion profile for random band matrices. Comm. Math. Phys. , 323(1):367--416, 2013

    2013

  21. [21]

    Ram\'irez, Benjamin Schlein, and Horng-Tzer Yau

    L\'aszl\'o Erd\"os, Sandrine P\'ech\'e, Jos\'e A. Ram\'irez, Benjamin Schlein, and Horng-Tzer Yau. Bulk universality for W igner matrices. Comm. Pure Appl. Math. , 63(7):895--925, 2010

    2010

  22. [22]

    Erd ¨os, V

    L \'a szl \'o Erd o s and Volodymyr Riabov. The zigzag strategy for random band matrices. arXiv preprint arXiv:2506.06441 , 2025

    work page arXiv 2025

  23. [23]

    Universality of random matrices and local relaxation flow

    L\'aszl\'o Erd\"os, Benjamin Schlein, and Horng-Tzer Yau. Universality of random matrices and local relaxation flow. Invent. Math. , 185(1):75--119, 2011

    2011

  24. [24]

    A dynamical approach to random matrix theory , volume 28 of Courant Lecture Notes in Mathematics

    L\'aszl\'o Erd \"o s and Horng-Tzer Yau. A dynamical approach to random matrix theory , volume 28 of Courant Lecture Notes in Mathematics . Courant Institute of Mathematical Sciences, New York; American Mathematical Society, Providence, RI, 2017

    2017

  25. [25]

    Feldheim and Sasha Sodin

    Ohad N. Feldheim and Sasha Sodin. A universality result for the smallest eigenvalues of certain sample covariance matrices. Geom. Funct. Anal. , 20(1):88--123, 2010

    2010

  26. [26]

    Localization-delocalization transition for a random block matrix model at the edge

    Jiaqi Fan, Bertrand Stone, Fan Yang, and Jun Yin. Localization-delocalization transition for a random block matrix model at the edge. arXiv preprint arXiv:2504.00512 , 2025

    work page arXiv 2025

  27. [27]

    A block reduction method for random band matrices with general variance profiles

    Jiaqi Fan, Fan Yang, and Jun Yin. A block reduction method for random band matrices with general variance profiles. arXiv preprint arXiv:2507.11945 , 2025

    work page arXiv 2025

  28. [28]

    Outliers for deformed inhomogeneous random matrices

    Ruohan Geng, Dang-Zheng Liu, and Guangyi Zou. Outliers for deformed inhomogeneous random matrices. arXiv preprint arXiv:2407.12182v2 , 2024

    work page arXiv 2024

  29. [29]

    Diffusion profile for random band matrices: a short proof

    Yukun He and Matteo Marcozzi. Diffusion profile for random band matrices: a short proof. J. Stat. Phys. , 177(4):666--716, 2019

    2019

  30. [30]

    Ramanujan Property and Edge Universality of Random Regular Graphs,

    Jiaoyang Huang, Theo McKenzie, and Horng-Tzer Yau. Ramanujan property and edge universality of random regular graphs. arXiv preprint arXiv:2412.20263 , 2024

    work page arXiv 2024

  31. [31]

    The E uler characteristic of the moduli space of curves

    John Harer and Don Zagier. The E uler characteristic of the moduli space of curves. Invent. Math. , 85(3):457--485, 1986

    1986

  32. [32]

    Universality of the local spacing distribution in certain ensembles of H ermitian W igner matrices

    Kurt Johansson. Universality of the local spacing distribution in certain ensembles of H ermitian W igner matrices. Comm. Math. Phys. , 215(3):683--705, 2001

    2001

  33. [33]

    Intersection theory on the moduli space of curves and the matrix A iry function

    Maxim Kontsevich. Intersection theory on the moduli space of curves and the matrix A iry function. Comm. Math. Phys. , 147(1):1--23, 1992

    1992

  34. [34]

    On the localization length of finite-volume random block Schr \"o dinger operators

    Steven Khang Truong , Fan Yang , and Jun Yin . On the localization length of finite-volume random block Schr \"o dinger operators . arXiv e-prints , page arXiv:2503.11382, March 2025

    work page arXiv 2025

  35. [35]

    The dimension-free structure of nonhomogeneous random matrices

    Rafa Lata a, Ramon van Handel, and Pierre Youssef. The dimension-free structure of nonhomogeneous random matrices. Invent. Math. , 214(3):1031--1080, 2018

    2018

  36. [36]

    Edge statistics of dyson brownian motion

    Benjamin Landon and Horng-Tzer Yau. Edge statistics of dyson brownian motion. arXiv preprint arXiv:1712.03881 , 2017

    work page arXiv 2017

  37. [37]

    Lando and Alexander K

    Sergei K. Lando and Alexander K. Zvonkin. Graphs on surfaces and their applications , volume 141 of Encyclopaedia of Mathematical Sciences . Springer-Verlag, Berlin, 2004. With an appendix by Don B. Zagier, Low-Dimensional Topology, II

    2004

  38. [38]

    Liu and G

    Dang-Zheng Liu and Guangyi Zou. Edge statistics for random band matrices. arXiv preprint arXiv:2401.00492v2 , 2024

    work page arXiv 2024

  39. [39]

    Liu and G

    Dang-Zheng Liu and Guangyi Zou. Edge universality for inhomogeneous random matrices. arXiv preprint arXiv:2508.17838v2 , 2025

    work page arXiv 2025

  40. [40]

    Duality of orthogonal and symplectic matrix integrals and quaternionic F eynman graphs

    Motohico Mulase and Andrew Waldron. Duality of orthogonal and symplectic matrix integrals and quaternionic F eynman graphs. Comm. Math. Phys. , 240(3):553--586, 2003

    2003

  41. [41]

    Random matrices and random permutations

    Andrei Okounkov. Random matrices and random permutations. Internat. Math. Res. Notices , 2000(20):1043--1095, 2000

    2000

  42. [42]

    On the W egner orbital model

    Ron Peled, Jeffrey Schenker, Mira Shamis, and Sasha Sodin. On the W egner orbital model. Int. Math. Res. Not. IMRN , 2019(4):1030--1058, 2019

    2019

  43. [43]

    On the second mixed moment of the characteristic polynomials of 1 D band matrices

    Tatyana Shcherbina. On the second mixed moment of the characteristic polynomials of 1 D band matrices. Comm. Math. Phys. , 328(1):45--82, 2014

    2014

  44. [44]

    J. G. Skellam. The frequency distribution of the difference between two P oisson variates belonging to different populations. J. Roy. Statist. Soc. (N.S.) , 109:296, 1946

    1946

  45. [45]

    The spectral edge of some random band matrices

    Sasha Sodin. The spectral edge of some random band matrices. Ann. of Math. (2) , 172(3):2223--2251, 2010

    2010

  46. [46]

    Universality at the edge of the spectrum in W igner random matrices

    Alexander Soshnikov. Universality at the edge of the spectrum in W igner random matrices. Comm. Math. Phys. , 207(3):697--733, 1999

    1999

  47. [47]

    Random banded and sparse matrices

    Thomas Spencer. Random banded and sparse matrices. In The O xford handbook of random matrix theory , pages 471--488. Oxford Univ. Press, Oxford, 2011

    2011

  48. [48]

    The approach to thermal equilibrium in quantized chaotic systems

    Mark Srednicki. The approach to thermal equilibrium in quantized chaotic systems. Journal of Physics A: Mathematical and General , 32(7):1163--1175, 1999

    1999

  49. [49]

    Universality for 1d random band matrices

    Mariya Shcherbina and Tatyana Shcherbina. Universality for 1d random band matrices. Comm. Math. Phys. , 385(2):667--716, 2021

    2021

  50. [50]

    A random matrix model towards the quantum chaos transition conjecture

    Bertrand Stone, Fan Yang, and Jun Yin. A random matrix model towards the quantum chaos transition conjecture. Comm. Math. Phys. , 406(4):Paper No. 85, 76, 2025

    2025

  51. [51]

    An introduction to matrix concentration inequalities

    Joel A Tropp. An introduction to matrix concentration inequalities. Foundations and Trends in Machine Learning , 8(1-2):1--230, 2015

    2015

  52. [52]

    Random matrices: universality of local eigenvalue statistics up to the edge

    Terence Tao and Van Vu. Random matrices: universality of local eigenvalue statistics up to the edge. Comm. Math. Phys. , 298(2):549--572, 2010

    2010

  53. [53]

    Random matrices: universality of local eigenvalue statistics

    Terence Tao and Van Vu. Random matrices: universality of local eigenvalue statistics. Acta Math. , 206(1):127--204, 2011

    2011

  54. [54]

    Outliers in spectrum of sparse W igner matrices

    Konstantin Tikhomirov and Pierre Youssef. Outliers in spectrum of sparse W igner matrices. Random Structures Algorithms , 58(3):517--605, 2021

    2021

  55. [55]

    Structured random matrices

    Ramon van Handel. Structured random matrices. In Convexity and concentration , volume 161 of IMA Vol. Math. Appl. , pages 107--156. Springer, New York, 2017

    2017

  56. [56]

    Disordered system with n orbitals per site: n= limit

    Franz J Wegner. Disordered system with n orbitals per site: n= limit. Physical Review B , 19(2):783, 1979

    1979

  57. [57]

    Eugene P. Wigner. Characteristic vectors of bordered matrices with infinite dimensions. Ann. of Math. (2) , 62:548--564, 1955

    1955

  58. [58]

    Random matrices in physics

    Eugene P Wigner. Random matrices in physics. SIAM review , 9(1):1--23, 1967

    1967

  59. [59]

    Bulk universality and quantum unique ergodicity for random band matrices in high dimensions

    Changji Xu, Fan Yang, Horng-Tzer Yau, and Jun Yin. Bulk universality and quantum unique ergodicity for random band matrices in high dimensions. Ann. Probab. , 52(3):765--837, 2024

    2024

  60. [60]

    Yang and J

    Fan Yang and Jun Yin. Delocalization of a general class of random block schr \"o dinger operators. arXiv preprint arXiv:2501.08608 , 2025

    work page arXiv 2025

  61. [61]

    Yang and J

    Fan Yang and Jun Yin. Delocalization of random band matrices at the edge. arXiv preprint arXiv:2505.11993 , 2025

    work page arXiv 2025

  62. [62]

    Delocalization of one-dimensional random band matrices.e-preprint arXiv:2501.01718

    Horng-Tzer Yau and Jun Yin. Delocalization of one-dimensional random band matrices. arXiv preprint arXiv:2501.01718 , 2025

    work page arXiv 2025

  63. [63]

    Yang, H.-T

    Fan Yang, Horng-Tzer Yau, and Jun Yin. Delocalization and quantum diffusion of random band matrices in high dimensions I : Self-energy renormalization. Preprint arXiv:2104.12048 , 2021

    work page arXiv 2021

  64. [64]

    Delocalization and quantum diffusion of random band matrices in high dimensions II : T -expansion

    Fan Yang, Horng-Tzer Yau, and Jun Yin. Delocalization and quantum diffusion of random band matrices in high dimensions II : T -expansion. Comm. Math. Phys. , 396(2):527--622, 2022

    2022