pith. sign in

arxiv: 2302.14823 · v3 · submitted 2023-02-28 · 🧮 math.PR

Full large deviation principles for the largest eigenvalue of sub-Gaussian Wigner matrices

Pith reviewed 2026-05-24 09:13 UTC · model grok-4.3

classification 🧮 math.PR
keywords large deviation principleslargest eigenvalueWigner matricessub-Gaussian entriesspherical spin glasseigenvector localizationrate function transition
0
0 comments X

The pith

The largest eigenvalue of sub-Gaussian Wigner matrices satisfies a large deviation principle that transitions from the GOE rate to a distribution-dependent one at the start of eigenvector localization.

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

The paper derives full large deviation principles for the largest eigenvalue of Wigner matrices whose entries follow sub-Gaussian distributions. It demonstrates that these deviations arise from a combination of a few large entries and collective behavior across many entries. The proof relies on approximating the tail probability at finite matrix size by an optimization problem over restricted annealed free energies in a spherical spin glass model. This method works whenever the log-Laplace transform of the entry law has bounded second derivative. The resulting rate function matches the Gaussian Orthogonal Ensemble exactly when the distribution is sharp sub-Gaussian, but switches to a non-universal form otherwise, at the deviation level where the eigenvector localizes.

Core claim

The upper tail of the largest eigenvalue λ1 in sub-Gaussian Wigner matrices can be approximated at finite N by an optimization over restricted annealed free energies of a spherical spin glass model. This approximation yields full large deviation principles provided the log-Laplace transform of the entry distribution has bounded second derivative. The rate function is that of the GOE if and only if the distribution is sharp sub-Gaussian; otherwise it transitions to a μ-dependent function precisely when the eigenvector localizes.

What carries the argument

finite-N approximation of the upper tail by an optimization problem involving restricted annealed free energies for a spherical spin glass model

If this is right

  • Precise upper-tail asymptotics hold via the spin glass optimization.
  • The rate function is universal (GOE) exactly for sharp sub-Gaussian distributions.
  • A transition in the rate function occurs for other sub-Gaussian distributions at the localization threshold for the eigenvector.
  • The argument covers a wider class of distributions than previous work requiring sharp sub-Gaussianity or symmetry.

Where Pith is reading between the lines

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

  • The technique of using restricted annealed free energies may apply to large deviations for other spectral quantities in random matrices.
  • Numerical verification of the transition point could be performed by sampling Wigner matrices with specific entry distributions like the uniform law.
  • The mixture of localized and delocalized effects suggests connections to outlier detection in covariance estimation with sub-Gaussian noise.

Load-bearing premise

The log-Laplace transform of the matrix entry distribution has bounded second derivative.

What would settle it

For a fixed non-sharp sub-Gaussian measure μ such as the uniform distribution on an interval, sample many large Wigner matrices and estimate the probability of λ1 exceeding a range of thresholds to check if the empirical rate function exhibits the predicted transition from GOE behavior.

Figures

Figures reproduced from arXiv: 2302.14823 by Alice Guionnet, Nicholas A. Cook, Raphael Ducatez.

Figure 1
Figure 1. Figure 1: Large deviations for λ1(H) for the case that µ is an equal mixture of the centered Gaussian of variance 2 and the Dirac mass δ0 (the case p = 1 2 of Example 2.4). (Computations were carried out with Mathematica.) Left: Numerically computed rate functions I γ (x) from (1.9) (green/blue) and I µ(x) from (2.34) (green/yellow), plotted for x ∈ [2, 3.5] (on a mesh of spacing .02). The rate functions match up to… view at source ↗
Figure 2
Figure 2. Figure 2: Plots of ψµ(t) = Λµ(t)/t2 for the standardized Bernoulli(p) measure (top), p-sparse Rademacher distribution (lower-left) and p-sparse Gaussian (lower￾right). In the latter two cases ψµ is symmetric. See Examples 2.2, 2.3, 2.4. The line ψγ(t) ≡ 1 2 for the Gaussian measure is plotted in red for reference. The standardized Bernoulli(p) measure is only sharp sub-Gaussian for p = 1 2 (the Rademacher case) and … view at source ↗
read the original abstract

We establish precise upper-tail asymptotics and large deviation principles for the rightmost eigenvalue $\lambda_1$ of Wigner matrices with sub-Gaussian entries. In contrast to the case of heavier tails, where deviations of $\lambda_1$ are due to the appearance of a few large entries, and the sharp sub-Gaussian case that is governed by the collective deviation of entries in a delocalized rank-one pattern, we show that the general sub-Gaussian case is determined by a mixture of localized and delocalized effects. Our key result is a finite-$N$ approximation for the upper tail of $\lambda_1$ by an optimization problem involving \emph{restricted annealed free energies} for a spherical spin glass model. This new type of argument allows us to derive full large deviation principles when the log-Laplace transform of the entries' distribution $\mu$ has bounded second derivative, whereas previous results required much more restrictive assumptions, namely sharp sub-Gaussianity and symmetry, or only covered certain ranges of deviations. We show that the sharp sub-Gaussian condition characterizes measures $\mu$ for which the rate function coincides with that of the Gaussian Orthogonal Ensemble (GOE). When $\mu$ is not sharp sub-Gaussian, at a certain distance from the bulk of the spectrum there is a transition from the GOE rate function to a non-universal rate function depending on $\mu$, and this transition coincides with the onset of a localization phenomenon for the associated eigenvector.

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

0 major / 0 minor

Summary. The manuscript claims to establish precise upper-tail asymptotics and full large deviation principles for the largest eigenvalue λ₁ of Wigner matrices with sub-Gaussian entries. The central result is a finite-N approximation of the upper tail via an optimization problem over restricted annealed free energies of a spherical spin glass model. This approach yields the LDP when the log-Laplace transform of the entry distribution μ has bounded second derivative, extending prior results that required sharp sub-Gaussianity and symmetry or covered only limited deviation ranges. The paper further shows that the sharp sub-Gaussian condition makes the rate function coincide with the GOE case, while non-sharp cases exhibit a transition to a μ-dependent rate function at a point coinciding with the onset of eigenvector localization.

Significance. If the finite-N approximation and resulting LDP hold, the work would be significant as it provides a unified treatment of localized and delocalized contributions to large deviations in the sub-Gaussian regime, relaxing restrictive assumptions from earlier literature. The spin-glass optimization technique for finite-N control is a methodological strength that could extend to related problems in random matrices and disordered systems. The characterization of the GOE-to-non-universal transition is a concrete advance.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary of our results on full LDPs for the largest eigenvalue of sub-Gaussian Wigner matrices and for highlighting the significance of the finite-N approximation via restricted annealed free energies and the GOE-to-non-universal transition. The recommendation is 'uncertain' but the report contains no specific major comments requiring point-by-point response.

Circularity Check

0 steps flagged

No circularity in abstract; derivation rests on external spin-glass model

full rationale

Only the abstract is available. It describes a finite-N approximation of the upper tail of λ1 via an optimization over restricted annealed free energies of a spherical spin glass model, used to obtain full LDPs under a bounded-second-derivative condition on the log-Laplace transform. No equations, self-definitions, fitted parameters renamed as predictions, or self-citation chains appear in the text. The central argument is presented as relying on an external optimization problem rather than reducing to the paper's own inputs by construction. This is the most common honest finding when no load-bearing internal reduction can be exhibited.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no explicit list of free parameters, axioms, or invented entities; the work presumably relies on standard sub-Gaussian tail assumptions and spin-glass variational principles whose precise statements are not visible here.

pith-pipeline@v0.9.0 · 5778 in / 1072 out tokens · 27692 ms · 2026-05-24T09:13:48.902304+00:00 · methodology

discussion (0)

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

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Large deviations of crowding in finite $\beta$-ensembles

    math.PR 2026-05 unverdicted novelty 7.0

    Establishes large deviation principle with speed n² for the normalized count of points in bounded set U for finite β-ensembles on R and C under suitable boundary conditions on U.

  2. Beyond the Big Jump: A Perturbative Approach to Stretched-Exponential Processes

    cond-mat.stat-mech 2026-03 unverdicted novelty 7.0

    A perturbative series supplies explicit corrections to the big-jump approximation for sums of stretched-exponential random variables, describing the crossover to moderate deviations.

  3. Large deviations for the smallest eigenvalue of a deformed GOE with an outlier

    math.PR 2024-08 unverdicted novelty 7.0

    Establishes LDP for smallest eigenvalue of GOE plus diagonal outlier matrix, generalizing prior cases.

Reference graph

Works this paper leans on

86 extracted references · 86 canonical work pages · cited by 3 Pith papers

  1. [1]

    H., Erd ˝os, L., and Kr ¨uger, T

    Ajanki, O. H., Erd ˝os, L., and Kr ¨uger, T. Quadratic vector equations on complex upper half-plane. Mem. Amer. Math. Soc. 261 , 1261 (2019), v+133

  2. [2]

    Maximum of the Riemann zeta function on a short interval of the critical line.Comm

    Arguin, L.-P., Belius, D., Bourgade, P., Radziwi l l, M., and Soundararajan, K. Maximum of the Riemann zeta function on a short interval of the critical line.Comm. Pure Appl. Math. 72, 3 (2019), 500–535

  3. [3]

    Complexity of random smooth functions on the high-dimensional sphere

    Auffinger, A., and Ben Arous, G. Complexity of random smooth functions on the high-dimensional sphere. Ann. Probab. 41, 6 (2013), 4214–4247

  4. [4]

    Random matrices and complexity of spin glasses

    Auffinger, A., Ben Arous, G., and ˇCern´y, J. Random matrices and complexity of spin glasses. Com- mun. Pure Appl. Math. 66 , 2 (2013), 165–201

  5. [5]

    Large deviations of the empirical spectral measure of supercritical sparse Wigner matrices

    Augeri, F. Large deviations of the empirical spectral measure of supercritical sparse Wigner matrices. Preprint, arXiv:2401.11925. LARGE DEVIATIONS FOR THE LARGEST EIGENVALUE OF SUB-GAUSSIAN WIGNER MATRICES 99

  6. [6]

    Large deviations principle for the largest eigenvalue of Wigner matrices without Gaussian tails

    Augeri, F. Large deviations principle for the largest eigenvalue of Wigner matrices without Gaussian tails. Electron. J. Probab. 21 (2016), Paper No. 32, 49

  7. [7]

    On the large deviations of traces of random matrices

    Augeri, F. On the large deviations of traces of random matrices. Ann. Inst. H. Poincar´ e, Probab. Stat. 54, 4 (2018), 2239–2285

  8. [8]

    Nonlinear large deviation bounds with applications to Wigner matrices and sparse Erd¨ os-R´ enyi graphs

    Augeri, F. Nonlinear large deviation bounds with applications to Wigner matrices and sparse Erd¨ os-R´ enyi graphs. Ann. Probab. 48, 5 (2020), 2404–2448

  9. [9]

    Large deviations of the largest eigenvalue of supercritical sparse Wigner matrices

    Augeri, F., and Basak, A. Large deviations of the largest eigenvalue of supercritical sparse Wigner matrices. Preprint, arXiv:2304.13364

  10. [10]

    Large deviations for the largest eigenvalue of sub-Gaussian matrices

    Augeri, F., Guionnet, A., and Husson, J. Large deviations for the largest eigenvalue of sub-Gaussian matrices. Comm. Math. Phys. 383 , 2 (2021), 997–1050

  11. [11]

    Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices

    Baik, J., Ben Arous, G., and P ´ech´e, S. Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices. Ann. Probab. 33, 5 (2005), 1643–1697

  12. [12]

    Upper tail of the spectral radius of sparse Erd¨ os-R´ enyi graphs.Probab

    Basak, A. Upper tail of the spectral radius of sparse Erd¨ os-R´ enyi graphs.Probab. Theory Related Fields 187, 3-4 (2023), 885–947

  13. [13]

    Upper tail large deviations of regular subgraph counts in Erd˝ os-R´ enyi graphs in the full localized regime

    Basak, A., and Basu, R. Upper tail large deviations of regular subgraph counts in Erd˝ os-R´ enyi graphs in the full localized regime. Comm. Pure Appl. Math. 76 , 1 (2023), 3–72

  14. [14]

    Large deviation principles via spherical integrals

    Belinschi, S., Guionnet, A., and Huang, J. Large deviation principles via spherical integrals. arXiv 2004.07117 (2020)

  15. [15]

    Exponential growth of random determinants beyond invariance

    Ben Arous, G., Bourgade, P., and McKenna, B. Exponential growth of random determinants beyond invariance. Probab. Math. Phys. 3 , 4 (2022), 731–789

  16. [16]

    Aging of spherical spin glasses

    Ben Arous, G., Dembo, A., and Guionnet, A. Aging of spherical spin glasses. Probab. Theory Related Fields 120, 1 (2001), 1–67

  17. [17]

    Large deviations for Wigner’s law and Voiculescu’s non-commutative entropy

    Ben Arous, G., and Guionnet, A. Large deviations for Wigner’s law and Voiculescu’s non-commutative entropy. Probab. Theory Related Fields 108 , 4 (1997), 517–542

  18. [18]

    The landscape of the spiked tensor model

    Ben Arous, G., Mei, S., Montanari, A., and Nica, M. The landscape of the spiked tensor model. Commun. Pure Appl. Math 72 , 11 (2019), 2282–2330

  19. [19]

    Large deviations of the extreme eigenvalues of random deformations of matrices

    Benaych-Georges, F., Guionnet, A., and Ma ¨ıda, M. Large deviations of the extreme eigenvalues of random deformations of matrices. Probab. Theory Related Fields 154 , 3-4 (2012), 703–751

  20. [20]

    Local semicircle law for Wigner matrices

    Benaych-Georges, F., and Knowles, A. Local semicircle law for Wigner matrices. In Advanced topics in random matrices, vol. 53 of Panor. Synth` eses. Soc. Math. France, Paris, 2017, pp. 1–90

  21. [21]

    Large deviations for quadratic forms of stationary Gaussian processes

    Bercu, B., Gamboa, F., and Rouault, A. Large deviations for quadratic forms of stationary Gaussian processes. Stochastic Process. Appl. 71, 1 (1997), 75–90

  22. [22]

    B., Bhattacharya, S., and Ganguly, S

    Bhattacharya, B. B., Bhattacharya, S., and Ganguly, S. Spectral edge in sparse random graphs: Upper and lower tail large deviations. Ann. Probab. 49, 4 (2021), 1847–1885

  23. [23]

    B., and Ganguly, S

    Bhattacharya, B. B., and Ganguly, S. Upper tails for edge eigenvalues of random graphs. SIAM J. Discrete Math. 34 , 2 (2020), 1069–1083

  24. [24]

    Performance of statistical tests for single-source detection using random matrix theory

    Bianchi, P., Debbah, M., Ma ¨ıda, M., and Najim, J. Performance of statistical tests for single-source detection using random matrix theory. IEEE Trans. Inform. Theory 57 , 4 (2011), 2400–2419

  25. [25]

    Large deviations for the largest eigenvalues and eigenvectors of spiked Gaussian random matrices

    Biroli, G., and Guionnet, A. Large deviations for the largest eigenvalues and eigenvectors of spiked Gaussian random matrices. Electron. Commun. Probab. 25 (2020), Paper No. 70, 13

  26. [26]

    G., Chistyakov, G

    Bobkov, S. G., Chistyakov, G. P., and G ¨otze, F. Strictly subgaussian probability distributions. Elec- tron. J. Probab. 29 (2024), –

  27. [27]

    A large deviation principle for Wigner matrices without Gaussian tails

    Bordenave, C., and Caputo, P. A large deviation principle for Wigner matrices without Gaussian tails. Ann. Probab. 42, 6 (2014), 2454–2496

  28. [28]

    Sur la d´ etermination d’un syst` eme orthogonal complet dans un espace de Riemann sym´ etrique clos

    Cartan, E. Sur la d´ etermination d’un syst` eme orthogonal complet dans un espace de Riemann sym´ etrique clos. Rend. Circ. Mat. Palermo 53 (1929), 217–252

  29. [29]

    Nonlinear large deviations

    Chatterjee, S., and Dembo, A. Nonlinear large deviations. Adv. Math. 299 (2016), 396–450

  30. [30]

    Chatterjee, S., and R. S. Varadhan, S. Large deviations for random matrices. Comm Stoch Anal 6 , 1 (2012), 1–13

  31. [31]

    Chatterjee, S., and Varadhan, S. R. S.The large deviation principle for the Erd˝ os-R´ enyi random graph. European J. Combin. 32 , 7 (2011), 1000–1017

  32. [32]

    Chung, F. R. K. Spectral graph theory , vol. 92 of CBMS Regional Conference Series in Mathematics . Conference Board of the Mathematical Sciences, Washington, DC; by the American Mathematical Society, Providence, RI, 1997

  33. [33]

    Large deviations of subgraph counts for sparse Erd˝ os-R´ enyi graphs.Adv

    Cook, N., and Dembo, A. Large deviations of subgraph counts for sparse Erd˝ os-R´ enyi graphs.Adv. Math. 373 (2020), 107289, 53

  34. [34]

    Large deviations of extreme eigenvalues of random matrices

    Dean, D., and Majumdar, S. Large deviations of extreme eigenvalues of random matrices. Phys. Rev. Lett. 97, 16 (2006), 160201, 4

  35. [35]

    Large deviations techniques and applications , second ed., vol

    Dembo, A., and Zeitouni, O. Large deviations techniques and applications , second ed., vol. 38 of Applica- tions of Mathematics (New York) . Springer-Verlag, New York, 1998

  36. [36]

    Matrix optimization under random external fields

    Dembo, A., and Zeitouni, O. Matrix optimization under random external fields. J. Stat. Phys. 159 , 6 (2015), 1306–1326. 100 N. A. COOK, R. DUCATEZ, AND A. GUIONNET

  37. [37]

    Ducatez, R., Guionnet, A., and Husson, J.Large deviation principle for the largest eigenvalue of random matrices with a variance profile, 2024

  38. [38]

    Dyson, F. J. A Brownian-motion model for the eigenvalues of a random matrix. J. Math. Phys. 3 (1962), 1191–1198

  39. [39]

    Wegner estimate and level repulsion for Wigner random matrices

    Erd˝os, L., Schlein, B., and Yau, H.-T. Wegner estimate and level repulsion for Wigner random matrices. Int. Math. Res. Not. IMRN , 3 (2010), 436–479

  40. [40]

    A., Schlein, B., and Yau, H.-T

    Erd˝os, L., P ´ech´e, S., Ram ´ırez, J. A., Schlein, B., and Yau, H.-T. Bulk universality for Wigner matrices. Comm. Pure Appl. Math. 63 , 7 (2010), 895–925

  41. [41]

    Local semicircle law and complete delocalization for Wigner random matrices

    Erd˝os, L., Schlein, B., and Yau, H.-T. Local semicircle law and complete delocalization for Wigner random matrices. Comm. Math. Phys. 287 , 2 (2009), 641–655

  42. [42]

    Universality of random matrices and local relaxation flow.Invent

    Erd˝os, L., Schlein, B., and Yau, H.-T. Universality of random matrices and local relaxation flow.Invent. Math. 185, 1 (2011), 75–119

  43. [43]

    Forrester, P. J. The spectral edge of random matrix ensembles. Nuclear Phys. B 402 (1993), 709–728

  44. [44]

    The eigenvalues of random symmetric matrices

    F¨uredi, Z., and Koml ´os, J. The eigenvalues of random symmetric matrices. Combinatorica 1 (1981), 233–241

  45. [45]

    V., and Le Doussal, P

    Fyodorov, Y. V., and Le Doussal, P. Topology trivialization and large deviations for the minimum in the simplest random optimization. J. Stat. Phys. 154 , 1-2 (2014), 466–490

  46. [46]

    Spectral large deviations of sparse random matrices

    Ganguly, S., Hiesmayr, E., and Nam, K. Spectral large deviations of sparse random matrices. Preprint, arXiv:2206.06954

  47. [47]

    Large deviations for the largest eigenvalue of Gaussian networks with constant average degree

    Ganguly, S., and Nam, K. Large deviations for the largest eigenvalue of Gaussian networks with constant average degree. Probab. Theory Related Fields 184 , 3-4 (2022), 613–679

  48. [48]

    Large deviations for the largest eigenvalue of Rademacher matrices

    Guionnet, A., and Husson, J. Large deviations for the largest eigenvalue of Rademacher matrices. Ann. Probab. 48, 3 (2020), 1436–1465

  49. [49]

    A Fourier view on the R-transform and related asymptotics of spherical integrals

    Guionnet, A., and Ma ¨ıda, M. A Fourier view on the R-transform and related asymptotics of spherical integrals. J. Funct. Anal. 222 , 2 (2005), 435–490

  50. [50]

    Large deviations for the largest eigenvalue of the sum of two random matrices

    Guionnet, A., and Ma ¨ıda, M. Large deviations for the largest eigenvalue of the sum of two random matrices. Electron. J. Probab. 25 (2020), 24. Id/No 14

  51. [51]

    Concentration of the spectral measure for large matrices

    Guionnet, A., and Zeitouni, O. Concentration of the spectral measure for large matrices. Electron. Commun. Prob. 5 (2000), 119–136 (electronic)

  52. [52]

    Large deviations asymptotics for spherical integrals

    Guionnet, A., and Zeitouni, O. Large deviations asymptotics for spherical integrals. J. Funct. Anal. 188 (2002), 461–515

  53. [53]

    Upper tails via high moments and entropic stability

    Harel, M., Mousset, F., and Samotij, W. Upper tails via high moments and entropic stability. Duke Math. J. 171 , 10 (2022), 2089–2192

  54. [54]

    On dimension-dependent concentration for convex Lipschitz functions in product spaces

    Huang, H., and Tikhomirov, K. On dimension-dependent concentration for convex Lipschitz functions in product spaces. Electron. J. Probab. 28 (2023), Paper No. 63, 23

  55. [55]

    Large deviations for the largest eigenvalue of matrices with variance profiles.Electron

    Husson, J. Large deviations for the largest eigenvalue of matrices with variance profiles.Electron. J. Probab. 27 (2022), 44. Id/No 74

  56. [56]

    Large deviations for the largest eigenvalue of generalized sample covariance matrices

    Husson, J., and McKenna, B. Large deviations for the largest eigenvalue of generalized sample covariance matrices. Preprint, arXiv:2302.02847

  57. [57]

    On fluctuations of eigenvalues of random Hermitian matrices

    Johansson, K. On fluctuations of eigenvalues of random Hermitian matrices. Duke Math. J. 91 (1998), 151–204

  58. [58]

    Universality of the local spacing distribution in certain ensembles of Hermitian Wigner matrices

    Johansson, K. Universality of the local spacing distribution in certain ensembles of Hermitian Wigner matrices. Comm. Math. Phys. 215 (2001), 683–705

  59. [59]

    P., and Snaith, N

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

  60. [60]

    Uniform Hanson-Wright type concentration inequalities for un- bounded entries via the entropy method

    Klochkov, Y., and Zhivotovskiy, N. Uniform Hanson-Wright type concentration inequalities for un- bounded entries via the entropy method. Electron. J. Probab. 25 (2020), Paper No. 22, 30

  61. [61]

    Lacroix-A-Chez-Toine, B., Fyodorov, Y., and Pierre, L. D. Replica-symmetry breaking transitions in the large deviations of the ground-state of a spherical spin-glass. arXiv:2306.11927 (2023)

  62. [62]

    Almost optimal bulk regularity conditions in the CLT for Wigner matrices

    Landon, B., and Sosoe, P. Almost optimal bulk regularity conditions in the CLT for Wigner matrices. Preprint, arXiv:2204.03419

  63. [63]

    Lata la, R.Some estimates of norms of random matrices. Proc. Amer. Math. Soc. 133, 5 (2005), 1273–1282

  64. [64]

    Extremal spectral behavior of weighted random d-regular graphs

    Lee, J., and Nam, K. Extremal spectral behavior of weighted random d-regular graphs. Preprint, arXiv:2306.03479

  65. [65]

    Lubetzky, E., and Zhao, Y.On replica symmetry of large deviations in random graphs.Random Structures Algorithms 47, 1 (2015), 109–146

  66. [67]

    Large deviations for the largest eigenvalue of rank one deformations of Gaussian ensembles

    Ma¨ıda, M. Large deviations for the largest eigenvalue of rank one deformations of Gaussian ensembles. Electron. J. Probab. 12 (2007), 1131–1150

  67. [68]

    N., and Schehr, G

    Majumdar, S. N., and Schehr, G. Top eigenvalue of a random matrix: large deviations and third order phase transition. J. Stat. Mech. Theory Exp. 2014 , 1 (2014), 31. Id/No p01012

  68. [69]

    May, R. M. Will a large complex system be stable? Nature 238 (1972), 413–414. LARGE DEVIATIONS FOR THE LARGEST EIGENVALUE OF SUB-GAUSSIAN WIGNER MATRICES 101

  69. [70]

    Large deviations for extreme eigenvalues of deformed Wigner random matrices

    McKenna, B. Large deviations for extreme eigenvalues of deformed Wigner random matrices. Electron. J. Probab. 26 (2021), Paper No. 34, 37

  70. [71]

    Mehta, M. L. Random matrices, third ed., vol. 142 of Pure and Applied Mathematics (Amsterdam) . Else- vier/Academic Press, Amsterdam, 2004

  71. [72]

    Montgomery, H. L. The pair correlation of zeros of the zeta function. In Analytic number theory (Proc. Sympos. Pure Math., Vol. XXIV, St. Louis Univ., St. Louis, Mo., 1972) (1973), pp. 181–193

  72. [73]

    Large deviations of the free energy in diluted mean-field spin-glass

    Parisi, G., and Rizzo, T. Large deviations of the free energy in diluted mean-field spin-glass. J. Phys. A 43, 4 (2010), 045001, 18

  73. [74]

    Eigenvalue spectra of random matrices for neural networks

    Rajan, K., and Abbott, L. Eigenvalue spectra of random matrices for neural networks. Physical review letters 97, 18 (2006), 188104

  74. [75]

    Universality at the edge of the spectrum in Wigner random matrices

    Soshnikov, A. Universality at the edge of the spectrum in Wigner random matrices. Comm. Math. Phys. 207, 3 (1999), 697–733

  75. [76]

    A., and Teng, S

    Spielman, D. A., and Teng, S. H. Smooth analysis of algorithms. In Proceedings of the international congress of Mathematicians (Beijing 2002) , vol. I. Higher Ed. Press, Beijing, 2002, pp. 597–606

  76. [77]

    A new look at independence

    Talagrand, M. A new look at independence. Annals Probab. 24 (1996), 1–34

  77. [78]

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

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

  78. [79]

    A., and Widom, H

    Tracy, C. A., and Widom, H. Level spacing distributions and the Airy kernel. Commun. Math. Phys. 159 (1994), 151–174

  79. [80]

    High-dimensional probability, vol

    Vershynin, R. High-dimensional probability, vol. 47 of Cambridge Series in Statistical and Probabilistic Mathematics. Cambridge University Press, Cambridge, 2018. An introduction with applications in data science, With a foreword by Sara van de Geer

  80. [81]

    Large deviations of the maximum eigenvalue in Wishart random matrices

    Vivo, P., Majumdar, S., and Bohigas, O. Large deviations of the maximum eigenvalue in Wishart random matrices. J. Phys. A 40 , 16 (2007), 4317–4337

Showing first 80 references.