Pith. sign in

REVIEW 3 minor 37 references

Limit theorems for a strongly irreducible product of independent random matrices under optimal moment assumptions

T0 review · 0 major / 3 minor · reviewed 2026-05-24 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Under a contraction assumption on the matrix distribution, the log of the ratio of the top two singular values of random products escapes linearly to infinity with exponential large deviation bounds.

desk verdict The paper's grouping of i.i.d. factors into Cartan-aligned words gives a self-contained route to linear escape and exponential convergence without stationary measures. read the letter →

arxiv 2402.05751 v4 pith:BRLAYCDK submitted 2024-02-08 math.PR math.DS

Axel Péneau This is my paper
classification math.PRmath.DS
keywords randommatrixproductssingularvaluesLyapunovexponentslargedeviationsCartanprojectionstrongirreducibilitylimittheoremscontractionassumption
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

The paper establishes that for a probability distribution ν on square matrices satisfying a contraction assumption, the sequence log(s1/s2) of the product matrices escapes to infinity at a linear rate and obeys exponential large deviation inequalities below that rate. It further proves that the image of a generic line under the product and the maximal eigenspace both converge exponentially fast to one common random limiting line. These statements are derived by partitioning the independent factors into aligned random blocks in the Cartan decomposition, with explicit moment control, rather than invoking stationary measures or the prior existence of Lyapunov exponents. Additional L^p integrability on a norm function yields uniform integrability of the logarithmic distance from the limit line to any proper subspace, and for p=1 the normalized logarithms of matrix coefficients converge almost surely to the top Lyapunov exponent.

What carries the argument

Grouping i.i.d. matrix factors into i.i.d. random words aligned in the Cartan decomposition with explicit moment control

What would settle it

Exhibiting a distribution ν obeying the contraction condition for which log(s1/s2(γ_n)) stays bounded for infinitely many n, or for which the image line and maximal eigenspace converge to distinct limits.

Watch

Extended reading notes

Core claim

Under the contraction assumption on ν, the sequence (log s1/s2(γ_n bar))_n escapes to infinity linearly and satisfies exponential large deviations inequalities below its escape rate. The image of a generic line by γ_n bar as well as its eigenspace of maximal eigenvalue both converge to the same random line ℓ^∞ at an exponential speed. When ν is supported on invertible matrices and the push-forward N_*ν is L^p, -log d(ℓ^∞, H) is uniformly L^p for every proper subspace H; for p=1 the rescaled logarithm of each coefficient of γ_n bar converges almost surely to the top Lyapunov exponent. The proofs proceed by grouping the i.i.d. factors into i.i.d. random words aligned in the Cartan projection,,

Load-bearing premise

The contraction assumption on the probability distribution ν.

Editorial extensions

If this is right

  • The top Lyapunov exponent is simple in the extended sense given by linear escape plus large deviations.
  • Exponential convergence holds for both the projected line and the maximal eigenspace to the same random limit.
  • Under L^p integrability the logarithmic distance from ℓ^∞ to any proper subspace is uniformly integrable.
  • For p=1 the normalized entrywise logarithms converge almost surely to the top Lyapunov exponent.

Reading between the lines

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

  • The block-grouping technique may extend to other asymptotic regimes for matrix products where stationary measures are unavailable.
  • Strong irreducibility plus contraction may suffice for many limit theorems even when matrices are not assumed invertible.
  • Numerical simulation of products under the contraction condition could directly test whether the observed escape rate of log(s1/s2) matches the large-deviation bound.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

0 major / 3 minor

Summary. The paper proves limit theorems for the random product of i.i.d. matrices drawn from a distribution ν on d×d matrices over a local field. Under a contraction assumption on ν, log(s1/s2(γ̄_n)) escapes linearly to +∞ and obeys exponential large-deviation bounds below its rate; the image of a generic line and the maximal eigenspace both converge exponentially to the same random line ℓ^∞. Under the additional assumption that N_*ν is L^p for N(g)=log‖g‖‖g^{-1}‖, -log dist(ℓ^∞,H) is uniformly L^p for proper subspaces H; for p=1 the normalized logarithms of the matrix entries converge a.s. to the top Lyapunov exponent. All proofs proceed by grouping the i.i.d. factors into Cartan-aligned random words with explicit moment bounds and avoid any appeal to stationary measures or pre-existing Lyapunov exponents.

Significance. If the proofs are complete, the work is significant: it obtains the stated limit theorems under moment assumptions that appear optimal, supplies explicit large-deviation and convergence rates, and introduces a self-contained grouping construction that bypasses the usual stationary-measure machinery. The explicit moment control and the fact that the method is parameter-free are concrete strengths that could be reused in related settings.

minor comments (3)
  1. [Abstract / Introduction] The title refers to a 'strongly irreducible' product, yet the abstract only mentions a 'contraction assumption'; a short paragraph relating the two notions should appear in the introduction.
  2. [Notation section] The notation γ̄_n for the partial product is introduced in the abstract; verify that the same overline convention is used uniformly in all statements of the main theorems.
  3. [Statement of Theorem on a.s. convergence] The final a.s. convergence statement invokes the top Lyapunov exponent; a brief remark confirming that its existence has already been established by the preceding escape-rate argument would remove any possible ambiguity.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their careful reading, positive assessment of the significance of the results, and recommendation of minor revision. The referee's summary accurately captures the main contributions. Since no specific major comments or requested changes are listed in the report, we interpret the minor revision as pertaining to any editorial or typographical matters that may arise during production. We are happy to incorporate such adjustments in the revised version.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity identified

full rationale

The derivation proceeds from an external contraction assumption on the input measure ν, using an explicit grouping construction on i.i.d. factors to produce aligned random words in the Cartan decomposition together with moment bounds; the linear escape rate for log(s1/s2) and the exponential convergence statements are obtained directly from this construction without invoking stationary measures or pre-existing Lyapunov exponents. No step reduces a claimed result to a quantity defined in terms of itself, renames a fitted parameter as a prediction, or relies on a load-bearing self-citation whose content is itself unverified. The central claims therefore retain independent content relative to the stated hypotheses.

Assumptions & free parameters 0 free parameters · 1 assumptions · 0 invented entities

The central claims rest on the contraction assumption on ν and standard background results from the theory of random walks on groups and Cartan decomposition; no free parameters or new entities are introduced.

assumptions (1)
  • domain assumption Contraction assumption on the probability distribution ν
    Invoked to guarantee linear escape of log(s1/s2) and exponential convergence; appears in the first paragraph of the abstract.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Limit theorems for a strongly irreducible product of independent random matrices under optimal moment assumptions." pith.science (2026). https://pith.science/paper/BRLAYCDK

@misc{pith2026240205751,
  author       = {Pith},
  title        = {Pith review of: Limit theorems for a strongly irreducible product of independent random matrices under optimal moment assumptions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BRLAYCDK}},
  note         = {Machine review of arXiv:2402.05751}
}
abstract

Let $\nu$ be a probability distribution over the semi-group of square matrices of size $d \ge 2$ over a locally compact field $\mathbb{K}$, \textit{e.g.} $\mathbb{R}$. We consider the random walk $\overline{\gamma}_n := \gamma_0\cdots\gamma_{n-1}$ for $(\gamma_k)_{k \in \mathbb{N}}$ independent of law $\nu$. Let $s_1 \ge s_2 \ge \dots \ge s_d$ be the singular values given by the Cartan projection. Under a contraction assumption on $\nu$, we show that $(\log\frac{s_1}{s_2}(\overline{\gamma}_n))_{n \in\mathbb{N}}$, escapes to infinity linearly and satisfies exponential large deviations inequalities below its escape rate. This extends the notion of simplicity of the top Lyapunov exponent. We also show that the image of a generic line by $\overline{\gamma}_n$ as well as its eigenspace of maximal eigenvalue both converge to the same random line $\ell^\infty$ at an exponential speed. If we moreover assume that $\nu$ is supported on the group of invertible matrices and that the push-forward distribution $N_*\nu$ is $\mathrm{L}^p$ for $N: g \mapsto\log\|g\|\|g^{-1}\|$ and for some $p > 0$, then we show that $- \log\mathrm{d}(\ell^\infty, H)$ is uniformly $\mathrm{L}^p$ for all proper subspace $H \subset \mathbb{R}^d$. For $p = 1$, we moreover show that the rescaled logarithm of each coefficient of $\overline{\gamma}_n$ almost surely converges to the top Lyapunov exponent. To prove these results, we do not rely on the existence of the stationary measure nor on the existence of the Lyapunov exponents. Instead we describe an effective way to group the i.i.d. factors into i.i.d. random words that are somehow aligned in the Cartan decomposition. We moreover have an explicit control over the moments.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    Abels, G

    H. Abels, G. A. Margulis, and G. A. Soifer. Semigroups containing proximal linear maps. Israel Journal of Mathematics , 91(1):1--30, 1995

    work page 1995

  2. [2]

    Random subgroups of linear groups are free

    Richard Aoun. Random subgroups of linear groups are free . Duke Mathematical Journal , 160(1):117 -- 173, 2011

    work page 2011

  3. [3]

    The central limit theorem for eigenvalues, 2020

    Richard Aoun. The central limit theorem for eigenvalues, 2020

    work page 2020

  4. [4]

    Law of large numbers for the spectral radius of random matrix products, 2021

    Richard Aoun and Cagri Sert. Law of large numbers for the spectral radius of random matrix products, 2021

    work page 2021

  5. [5]

    Limit theorems for non-commutative operations

    Richard Bellman. Limit theorems for non-commutative operations. I. Duke Mathematical Journal , 21(3):491 -- 500, 1954

    work page 1954

  6. [6]

    Y. Benoist. Propri \'e t \'e s asymptotiques des groupes lin \'e aires. Geometric & Functional Analysis GAFA , 7(1):1--47, Mar 1997

    work page 1997

  7. [7]

    o dinger operators , volume 8 of Prog. Probab. Stat. Birkh \

    Philippe Bougerol and Jean Lacroix. Products of random matrices with applications to Schr \"o dinger operators , volume 8 of Prog. Probab. Stat. Birkh \"a user, Boston, MA, 1985

    work page 1985

  8. [8]

    large deviations for random walks on gromov-hyperbolic spaces, 2020

    Adrien Boulanger, Pierre Mathieu, Cagri Sert, and Alessandro Sisto. large deviations for random walks on gromov-hyperbolic spaces, 2020

    work page 2020

Show all 37 references
  1. [9]

    Random walks on reductive groups , volume 62 of Ergebnisse der Mathematik und ihrer Grenzgebiete

    Yves Benoist and Jean-Fran c ois Quint. Random walks on reductive groups , volume 62 of Ergebnisse der Mathematik und ihrer Grenzgebiete. 3. Folge. A Series of Modern Surveys in Mathematics [Results in Mathematics and Related Areas. 3rd Series. A Series of Modern Surveys in Ma...

    2016

  2. [10]

    Central limit theorem for linear groups

    Yves Benoist and Jean-Fran c ois Quint. Central limit theorem for linear groups . The Annals of Probability , 44(2):1308 -- 1340, 2016

    2016

  3. [11]

    Central limit theorem on hyperbolic groups

    Yves Benoist and Jean-François Quint. Central limit theorem on hyperbolic groups. Izvestiya: Mathematics , 80:3 -- 23, 2016

    2016

  4. [12]

    James W. Cannon. The combinatorial structure of cocompact discrete hyperbolic groups. 16:123--148, 1984

    1984

  5. [13]

    Limit theorems for the left random walk on gld (r), 2016

    Christophe Cuny, Jerome Dedecker, and Christophe Jan. Limit theorems for the left random walk on gld (r), 2016

    2016

  6. [14]

    On the koml\'os, major and tusn\'ady strong approximation for some classes of random iterates, 2017

    Christophe Cuny, Jérôme Dedecker, and Florence Merlevède. On the koml\'os, major and tusn\'ady strong approximation for some classes of random iterates, 2017

    2017

  7. [15]

    Limit theorems for iid products of positive matrices, 2023

    C Cuny, J Dedecker, and F Merlevède. Limit theorems for iid products of positive matrices, 2023

    2023

  8. [16]

    The poisson boundary of hyperbolic groups without moment conditions, 2022

    Kunal Chawla, Behrang Forghani, Joshua Frisch, and Giulio Tiozzo. The poisson boundary of hyperbolic groups without moment conditions, 2022

    2022

  9. [17]

    Random walks and contracting elements i: Deviation inequality and limit laws, 2024

    Inhyeok Choi. Random walks and contracting elements i: Deviation inequality and limit laws, 2024

    2024

  10. [18]

    Random walks and contracting elements ii: Translation length and quasi-isometric embedding, 2024

    Inhyeok Choi. Random walks and contracting elements ii: Translation length and quasi-isometric embedding, 2024

    2024

  11. [19]

    Random walks and contracting elements iii: Outer space and outer automorphism group, 2024

    Inhyeok Choi. Random walks and contracting elements iii: Outer space and outer automorphism group, 2024

    2024

  12. [20]

    Furstenberg and H

    H. Furstenberg and H. Kesten. Products of random matrices. The Annals of Mathematical Statistics , 31(2):457--469, 1960

    1960

  13. [21]

    Boundary theory and stochastic processes on homogeneous spaces

    Harry Furstenberg. Boundary theory and stochastic processes on homogeneous spaces. Harmonic analysis on homogeneous spaces , 26:193--229, 1973

    1973

  14. [22]

    Goldsheid and G

    Ilya Ya. Goldsheid and G. A. Margulis. Lyapunov indices of a product of random matrices. Russian Mathematical Surveys , 44:11--71, 1989

    1989

  15. [23]

    Exponential bounds for random walks on hyperbolic spaces without moment conditions

    Sébastien Gouëzel. Exponential bounds for random walks on hyperbolic spaces without moment conditions. Tunisian Journal of Mathematics , 4(4):635–671, December 2022

    2022

  16. [24]

    A zero-one law for invariant measures and a local limit theorem for coefficients of random walks on the general linear group, 2020

    Ion Grama, Jean-François Quint, and Hui Xiao. A zero-one law for invariant measures and a local limit theorem for coefficients of random walks on the general linear group, 2020

    2020

  17. [25]

    Frontiere de furstenberg, propri \'e t \'e s de contraction et th \'e oremes de convergence

    Yves Guivarc'h and Albert Raugi. Frontiere de furstenberg, propri \'e t \'e s de contraction et th \'e oremes de convergence. Zeitschrift f \"u r Wahrscheinlichkeitstheorie und Verwandte Gebiete , 69:187--242, 1985

    1985

  18. [26]

    Guivarc'h and A

    Y. Guivarc'h and A. Raugi. Products of random matrices: convergence theorems. In Random matrices and their applications ( B runswick, M aine, 1984) , volume 50 of Contemp. Math. , pages 31--54. Amer. Math. Soc., Providence, RI, 1986

    1984

  19. [27]

    Propri \'e t \'e s de contraction d'un semi-groupe de matrices inversibles

    Yves Guivarc'h and Albert Raugi. Propri \'e t \'e s de contraction d'un semi-groupe de matrices inversibles. coefficients de liapunoff d'un produit de matrices al \'e atoires ind \'e pendantes. Israel Journal of Mathematics , 65(2):165--196, 1989

    1989

  20. [28]

    Guivarc'h

    A. Guivarc'h. Quelques proprietes asymptotiques des produits de matrices aleatoires. In P. L. Hennequin, editor, Ecole d'Et \'e de Probabilit \'e s de Saint-Flour VIII-1978 , pages 177--250, Berlin, Heidelberg, 1980. Springer Berlin Heidelberg

    1978

  21. [29]

    H. Hennion. Limit theorems for products of positive random matrices . The Annals of Probability , 25(4):1545 -- 1587, 1997

    1997

  22. [30]

    J. F. C. Kingman. The ergodic theory of subadditive stochastic processes. J. Roy. Statist. Soc. Ser. B , 30:499--510, 1968

    1968

  23. [31]

    Convergence in distribution of products of random matrices

    Harry Kesten and Frank Spitzer. Convergence in distribution of products of random matrices. Zeitschrift f \"u r Wahrscheinlichkeitstheorie und Verwandte Gebiete , 67(4):363--386, 1984

    1984

  24. [32]

    Theoremes limites pour les produits de matrices aleatoires

    Emile Le Page. Theoremes limites pour les produits de matrices aleatoires. In Herbert Heyer, editor, Probability Measures on Groups , pages 258--303, Berlin, Heidelberg, 1982. Springer Berlin Heidelberg

    1982

  25. [33]

    Mathieu and A

    P. Mathieu and A. Sisto. Deviation inequalities for random walks. Duke Mathematical Journal , 169(5), Apr 2020

    2020

  26. [34]

    Convergence in distribution of products of random matrices: A semigroup approach

    Arunava Mukherjea. Convergence in distribution of products of random matrices: A semigroup approach. Transactions of the American Mathematical Society , 303(1):395--411, 1987

    1987

  27. [35]

    Large deviation principle for random matrix products, 2018

    Cagri Sert. Large deviation principle for random matrix products, 2018

    2018

  28. [36]

    Limit theorems for the coefficients of random walks on the general linear group, 2021

    Hui Xiao, Ion Grama, and Quansheng Liu. Limit theorems for the coefficients of random walks on the general linear group, 2021

    2021

  29. [37]

    Edgeworth expansion and large deviations for the coefficients of products of positive random matrices, 2022

    Hui Xiao, Ion Grama, and Quansheng Liu. Edgeworth expansion and large deviations for the coefficients of products of positive random matrices, 2022

    2022

Pith tools

Reviewed May 24, 2026 · model on record in the stance chip above.