Pith. sign in

REVIEW 5 minor 1 cited by

Self-affine sponges with random contractions

T0 review · 0 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A self-affine sponge with random contractions almost surely has Hausdorff and Minkowski dimension equal to $\min\{d,s_0\}$, defined by an expected-pressure equation, with no separation condition needed.

desk verdict Randomizing the contractions removes the separation condition for self-affine sponges and yields the expected dimension formula; the proof is intricate, and the explicit cost is the eventual-smoothness assumption (R4). read the letter →

arxiv 2505.04383 v2 pith:6U3MD6C7 submitted 2025-05-07 math.DS math.PR

Balázs Bárány , Antti Käenmäki , Micha{l} Rams This is my paper · ORCID
classification math.DSmath.PR MSC 28A8037C4560D05
keywords randomself-affinesetspongeHausdorffdimensionMinkowskiiteratedfunctionsystemsingularvaluetransversalityseparationcondition
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

Random self-affine sponges are fractals in $\mathbb{R}^d$ built from diagonal contraction matrices whose entries are drawn independently at each level of the construction. The paper proves that, almost surely, the Hausdorff and Minkowski dimensions of such a sponge coincide and equal $\min\{d,s_0\}$, where $s_0$ is the unique root of the expected-pressure equation $\max_\sigma \sum_{i=1}^N \mathbb{E}(\varphi_{s_0}^\sigma(A_i))=1$. The formula needs no separation condition: overlaps, including exact coincidences between different cylinder maps, do not lower the almost sure dimension. When $s_0>d$, the sponge has positive Lebesgue measure almost surely. The result extends the classical generic dimension formula for self-affine sets to a random, higher-dimensional, separation-free setting.

What carries the argument

The load-bearing objects are the coordinate-permuted singular value function $\varphi_s^\sigma(A)$ for diagonal matrices and its expected-pressure root $s_0$, defined by $\max_\sigma\sum_{i=1}^N \mathbb{E}(\varphi_{s_0}^\sigma(A_i))=1$. To prove the lower bound, the paper introduces finite-step subsystems $\Gamma_n$ whose words all contain prescribed blocks $\ell_k k_k$ in every coordinate, and constructs a random measure $\mu_n$ on them by a martingale limit, with cylinder weights proportional to $\varphi_{s_n}^\sigma(i)X^i$. The decisive probabilistic estimate is a transversality bound, Proposition 3.4, which controls the conditional probability that two distinct words map within distance $\varrho$: it is bounded by a product over coordinates of $\min\{1,C\varrho/|\alpha_{i\wedge j}^{(k)}|\}$. This estimate comes from the eventual smoothness of one log-contraction distribution in each coordinate, and it makes the energy integral of $\mu_n$ finite for every exponent below $s_n$, yielding the dimension lower bound.

What would settle it

Run the one-dimensional system of Example 1.3 with one random ratio uniform on $[1/3,1/2]$ and fixed ratios $1/3$ and $1/4$, and estimate the Hausdorff and box dimensions of deep prefixes; the theorem predicts both are almost surely $1$, the root of $\int_{1/3}^{1/2}6x^s\,dx+3^{-s}+4^{-s}=1$. A systematic gap between the measured dimensions and $s=1$, or a single realization with $\dim_H(X)<\dim_M(X)$, would falsify the no-separation claim. Equivalently, one may compute the conditional probability in Proposition 3.4 for a candidate distribution: if the bound $P(\alpha_{\ell_k}^{(k)}\in B(x,r))\le Cr$ is violated, e.g. for atomic log-contraction laws, the lower-bound proof cannot start.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.2: for a self-affine sponge with random contractions, almost surely $\dim_H(X)=\dim_M(X)=\min\{d,s_0\}$, and if $s_0>d$ then $L^d(X)>0$. Here $s_0$ is defined by taking the maximum over coordinate permutations $\sigma$ of the expected sum $\sum_i \mathbb{E}(\varphi_{s_0}^\sigma(A_i))$, where $\varphi_s^\sigma(A)$ is the singular value function adapted to a diagonal matrix. The proof obtains the upper bound by a covering argument and the lower bound by approximating the system with finite-step multiplicative subsystems and constructing a random measure on them. The no-separation feature is the main progress: previous results for diagonal sponges relied on exponential separation, while here the almost sure dimension is fixed by the distribution of the random contraction ratios alone.

Load-bearing premise

The theorem rests on assumption (R4): for every coordinate $k$, at least one of the random contraction ratios must have a log-distribution whose repeated self-convolution eventually has a continuous density. If every log-contraction distribution is atomic, the small-ball estimate $P(\alpha_{\ell_k}^{(k)}\in B(x,r))\le Cr$ fails and the transversality-based lower bound loses its main probabilistic handle.

Editorial extensions

If this is right

  • For any random sponge in the model, the almost sure dimension is determined by a closed-form expected-pressure equation, so $s_0$ can be evaluated numerically from the contraction-ratio distributions alone.
  • Overlaps, including exact overlaps between different words, do not change the almost sure dimension; Example 1.3 exhibits coincidences such as $f_{i2}\circ f_{i3}=f_{i3}\circ f_{i2}$ and still has dimension given by the formula.
  • When $s_0>d$, the random sponge has positive Lebesgue measure almost surely, so dimension saturation is accompanied by nonzero volume.
  • In dimension one the theorem reduces to the known almost sure dimension formula for uniformly random self-similar sets, and the same equation gives the similarity dimension.
  • The finite-step subsystem approximation shows that $\min\{d,s_0\}$ is approached from below by dimensions of increasingly refined random subsystems, making the formula stable under truncation.

Reading between the lines

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

  • An unstated consequence of the proof is that the only essential use of randomness is the small-ball estimate (3.3); any deterministic family of contraction ratios satisfying such a power-law estimate would satisfy the same dimension formula by the identical argument.
  • For the disjoint-rectangle version of the 4-corner construction, when $s_0>2$ the almost sure positive area should combine with the recursive placement to give non-empty interior almost surely, in analogy with the known one-dimensional results; the paper does not claim this.
  • We infer that the diagonal assumption is likely not essential: the coordinate-wise transversality argument suggests the same almost sure equality should hold for random self-affine sets with non-diagonal matrices when the singular-value distributions satisfy an analogous eventual smoothness condition.
  • A testable extension is to weaken (R4) to a Hölder-type small-ball condition on $\log\alpha_{\ell_k}^{(k)}$; if the density estimate (3.3) survives with a power law, the whole lower-bound proof should go through unchanged.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 5 minor

Summary. The paper proves an almost-sure dimension formula for random self-affine sponges in R^d with diagonal matrices and random contraction rates, assuming independence, identical distribution, uniform bounds, and eventual smoothness of each coordinate's logarithmic contraction distribution (assumptions R1-R4). The main result, Theorem 1.2, states that almost surely dim_H(X)=dim_M(X)=min{d,s0}, where s0 is the unique solution of max_sigma sum_i E(phi_{s0}^sigma(A_i))=1, and that L^d(X)>0 almost surely when s0>d. The proof combines a covering argument for the upper bound, an n-step multiplicative system and martingale random measure for the lower bound, and a coordinate-wise transversality estimate, followed by a density argument for positive Lebesgue measure. Several examples illustrate the scope, including a one-dimensional random self-similar set with overlaps and random 4-corner sets in the plane.

Significance. If the result is correct, it is a substantial contribution to the dimension theory of self-affine sets: it removes all separation conditions in the random sponge setting and extends Koivusalo's one-dimensional formula to higher-dimensional diagonal systems. The theorem is precisely stated and the dimension parameter s0 is defined by an explicit expectation equation, so the formula is checkable and has no free fitted parameters. The proof is structurally sound: the upper bound is a clean covering argument, and the lower bound follows the Koivusalo template with a martingale measure and coordinate transversality. The eventual smoothness condition (R4) is the crucial hypothesis; it is stated explicitly and is used exactly where the transversality estimate needs a quantitative small-ball probability, so this is a legitimate assumption rather than a hidden gap. The examples, including systems with exact overlaps, are useful and demonstrate that the assumptions are natural.

minor comments (5)
  1. [§3.1, Proposition 3.3] When p>1, Lemma 3.2 is cited for the lower bound on |π_{i∧j·i'|k}(σ^{k+1}i')−t_ℓ|, but σ^{k+1}i' begins with the repeated letter ℓ rather than with the cylinder used in Lemma 3.2; the bound is nevertheless true after iterating f_ℓ and replacing c by α^{p−1}c, so the proof should state this iteration explicitly.
  2. [§4.1, Proposition 4.1] The displayed estimate P(A_n α^{ns}>1)≤P(α<∑_{i∈C_n}ψ^s(i)) is not literally a consequence of the preceding covering bound; tracking the constants from the covering ratio, including the factor 2 and the relationship between A_n and ψ^s(i), gives the same Borel-Cantelli conclusion, but the chain as written should be rewritten.
  3. [§4.2, Proposition 4.2] The step in which µ_n([hip])µ_n([hjq]) is replaced by φ^{sn}_σ(hip)φ^{sn}_σ(hjq) and the X-factors are removed is compressed: the proof should specify the σ-algebra F_{n+m+1}, the conditional independence of X_{hip} and X_{hjq}, and the measurability of the indicator event. In addition, the notation π^k_h(ip) for finite words ip should be disambiguated from projections of infinite words.
  4. [§1.1, Example 1.5] In the definition of f_{i2}, the fourth coordinate term appears as α^{(4)}_{i4}; this looks like a typo and should be α^{(4)}_i.
  5. [§4.2, Proposition 4.2] The displayed formulas use s both for the dimension parameter and for the fixed exponent in constants such as (α/α)^{s(p+p')}; using s_n or t consistently would remove ambiguity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the dimension formula is derived from the stated random contraction model, with s0 determined by the laws of the random matrices before any dimension computation.

full rationale

The paper defines s0 as the unique solution of max_sigma sum_i E(phi_{s0}^sigma(A_i)) = 1 using only the distributions of the random diagonal matrices, before the dimension theorem is stated. The proof does not fit any parameter to the self-affine set X nor to its dimension; the upper bound in Proposition 4.1 is a direct covering estimate with Borel-Cantelli, and the lower bound in Proposition 4.2 constructs a random measure whose expected energy is controlled through the transversality estimate of Proposition 3.4, which rests on the eventual smoothness assumption (R4). No step reduces to the conclusion by definition, and no fitted quantity is renamed as a prediction. The self-citations appearing in the introduction, such as [5] and [6], are contextual references to prior results and are not invoked as load-bearing support in the proof of Theorem 1.2. The result is thus self-contained in the relevant sense: it proves, rather than assumes, the equality dim_H(X)=dim_M(X)=min{d,s0}.

Assumptions & free parameters 0 free parameters · 8 assumptions · 0 invented entities

The central claim is derived from explicit stochastic assumptions on the diagonal contractions. The only non-standard input is (R4), eventual smoothness of one log-contraction distribution per coordinate; it is stated upfront and is needed for transversality. No free parameters are fitted to data and no new physical or mathematical entities are postulated beyond the random IFS itself.

assumptions (8)
  • standard math Existence and uniqueness of the attractor X for a finite contracting iterated function system (Hutchinson's theorem).
    Invoked at the start of Section 1 to define X as the unique compact set satisfying X = union_i phi_i(X).
  • standard math Doob's L2 martingale convergence theorem.
    Used in Theorem 2.1 to obtain the L2 limit X^i that defines the random measure mu_n.
  • standard math Hahn-Kolmogorov extension theorem.
    Used in Theorem 2.1 to extend cylinder-set values to a Borel probability measure on the symbolic space.
  • standard math Mattila [24, Theorem 8.7]: finite s-energy implies Hausdorff dimension at least s.
    Used in the final step of Proposition 4.2 to convert the finite energy estimate into the lower bound on dim_H(X).
  • standard math Mattila [24, Theorem 2.12(iii)]: finite lower local density implies absolute continuity.
    Used in Proposition 4.3 to conclude L^d(X)>0 from the finiteness of the expected lower density of the projected measure.
  • domain assumption (R1)-(R3): tree-wise independence, identical distribution, and uniform bounds alpha <= |alpha_i^{(k)}| <= alpha for all nodes and all coordinates.
    These explicit model assumptions define the random contractions and are used throughout the expectation, martingale, covering, and transversality estimates.
  • domain assumption (R4): for each coordinate k, log alpha_{l_k}^{(k)} has an eventually smooth distribution in the sense of (1.2).
    This is the key quantitative input behind the density estimate (3.3) and the transversality Propositions 3.3 and 3.4; without it the lower-bound proof breaks.
  • domain assumption For each coordinate k, the translation values t_i dot e_k are not all equal.
    Guarantees every coordinate projection is a nontrivial random self-similar set, a prerequisite for the one-dimensional transversality argument in Section 3.1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Self-affine sponges with random contractions." pith.science (2026). https://pith.science/paper/6U3MD6C7

@misc{pith2026250504383,
  author       = {Pith},
  title        = {Pith review of: Self-affine sponges with random contractions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6U3MD6C7}},
  note         = {Machine review of arXiv:2505.04383}
}
abstract

We compute the almost sure Hausdorff dimension of random self-affine sponges in $\mathbb{R}^d$ without imposing any separation conditions. In this context, randomness arises from the matrices in the defining semigroup, which are random yet the corresponding affine maps share a fixed point.

Figures

Figures reproduced from arXiv: 2505.04383 by the authors.

Figure 1
Figure 1. Illustration for the self-similar set with random contractions defined in Example 1.3. The theorem generalizes Koivusalo’s result [23, Theorem 2.2], incorporating several key concepts from its proof. Establishing s0 as an upper bound for the almost sure upper Minkowski dimension is achieved through a simple covering argument. To derive a lower bound for the almost sure Hausdorff dimension, we employ n-step multiplic… view at source ↗
Figure 2
Figure 2. The first and second iterates of the RIFS defined in (1.5), and the random 4-corner set X of Example 1.4. for some i ̸= j. The resulting self-similar set X with random contractions clearly satisfies assumptions (R1)–(R4). Notably, each iteration involves only one random cylinder, distinguishing this framework from previously studied settings and extending the scope of existing results. Consider an example where the … view at source ↗
Figure 3
Figure 3. The first and second iterates of the RIFS defined in (1.6), and the random 4-corner set X of Example 1.5. Define the affine maps by setting fi1(x, y) = (α (1) i1 x, α (2) i1 y), fi2(x, y) = (α (1) i2 x, α (2) i2 y + (1 − α (2) i2 )), fi3(x, y) = (α (1) i3 x + (1 − α (1) i3 ), α (2) i3 y + (1 − α (2) i3 )), fi4(x, y) = (α (1) i4 x + (1 − α (1) i4 ), α (2) i4 y), (1.5) for all i ∈ {1, 2, 3, 4} ∗ and (x, y) ∈ R 2 . The… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. On the Fourier transform of random Bernoulli convolutions

    math.DS 2025-07 accept novelty 7.0 of 10

    For random Bernoulli convolutions, the Fourier transform is in L^1 almost surely whenever λ_g > 2/π, giving absolute continuity and non-empty interior; polynomial Fourier decay holds for every λ_g.

Reference graph

Works this paper leans on

32 extracted references · 21 canonical work pages · cited by 1 Pith paper

  1. [1]

    Algom, F

    A. Algom, F. R. Hertz, and Z. Wang. Polynomial Fourier decay and a cocycle version of Dolgopyat’s method for self-conformal measures. preprint, available at arXiv:2306.01275, 2024

    arXiv 2024

  2. [2]

    Baker and A

    S. Baker and A. Banaji. Polynomial Fourier decay for fractal measures and their pushforwards. Math. Ann., 392(1):209–261, 2025

    work page 2025

  3. [3]

    B´ ar´ any

    B. B´ ar´ any. Dimension of the generalized 4-corner set and its projections.Ergodic Theory Dynam. Systems, 32(4):1190–1215, 2012

    work page 2012

  4. [4]

    B´ ar´ any, M

    B. B´ ar´ any, M. Hochman, and A. Rapaport. Hausdorff dimension of planar self-affine sets and measures. Invent. Math., 216(3):601–659, 2019

    work page 2019

  5. [5]

    B´ ar´ any, A

    B. B´ ar´ any, A. K¨ aenm¨ aki, and H. Koivusalo. Dimension of self-affine sets for fixed translation vectors. J. Lond. Math. Soc. (2) , 98(1):223–252, 2018

    work page 2018

  6. [6]

    B´ ar´ any and M

    B. B´ ar´ any and M. Rams. Smoothness of random self-similar measures on the line and the existence of interior points. preprint, available at arXiv:2412.06008, 2024

    arXiv 2024

  7. [7]

    T. Bedford. Crinkly curves, markov partitions and box dimensions in self-similar sets. Ph.D disserta- tion, University of Warwick, 1984

    work page 1984

  8. [8]

    Dekking, K

    M. Dekking, K. Simon, and B. Sz´ ekely. The algebraic difference of two random Cantor sets: the Larsson family. Ann. Probab., 39(2):549–586, 2011

    work page 2011

Show all 32 references
  1. [9]

    Dekking, K

    M. Dekking, K. Simon, B. Sz´ ekely, and N. Szekeres. The interior of randomly perturbed self-similar sets on the line. Adv. Math., 448:Paper No. 109724, 43, 2024

    2024

  2. [10]

    J. L. Doob. Stochastic processes. Wiley Classics Library. John Wiley & Sons, Inc., New York, 1990. Reprint of the 1953 original, A Wiley-Interscience Publication

    1990

  3. [11]

    G. A. Edgar. Fractal dimension of self-affine sets: some examples. Number 28, pages 341–358. 1992. Measure theory (Oberwolfach, 1990)

    1992

  4. [12]

    Falconer

    K. Falconer. Fractal geometry. John Wiley & Sons, Ltd., Chichester, third edition, 2014. Mathematical foundations and applications

    2014

  5. [13]

    K. J. Falconer. Random fractals. Math. Proc. Cambridge Philos. Soc. , 100(3):559–582, 1986

    1986

  6. [14]

    K. J. Falconer. The Hausdorff dimension of self-affine fractals. Math. Proc. Cambridge Philos. Soc. , 103(2):339–350, 1988. 24 BAL ´AZS B ´AR´ANY, ANTTI K ¨AENM¨AKI, AND MICHA L RAMS

    1988

  7. [15]

    Z. Feng. Dimension of diagonal self-affine measures with exponentially separated projections. Preprint, available at arXiv:2501.17378, 2025

    2025 arXiv

  8. [16]

    S. Graf. Statistically self-similar fractals. Probab. Theory Related Fields, 74(3):357–392, 1987

    1987

  9. [17]

    S. Graf, R. D. Mauldin, and S. C. Williams. The exact Hausdorff dimension in random recursive constructions. Mem. Amer. Math. Soc. , 71(381):x+121, 1988

    1988

  10. [18]

    Gu and J

    Y. Gu and J. J. Miao. Generalized q-dimensions of measures on nonautonomous fractals. preprint, available at arXiv:2411.17298, 2024

    2024 arXiv

  11. [19]

    Hochman and A

    M. Hochman and A. Rapaport. Hausdorff dimension of planar self-affine sets and measures with overlaps. J. Eur. Math. Soc. (JEMS) , 24(7):2361–2441, 2022

    2022

  12. [20]

    J. E. Hutchinson. Fractals and self-similarity. Indiana Univ. Math. J. , 30(5):713–747, 1981

    1981

  13. [21]

    Jordan and N

    T. Jordan and N. Jurga. Self-affine sets with non-compactly supported random perturbations. Ann. Acad. Sci. Fenn. Math., 39(2):771–785, 2014

    2014

  14. [22]

    Jordan, M

    T. Jordan, M. Pollicott, and K. Simon. Hausdorff dimension for randomly perturbed self affine attractors. Comm. Math. Phys. , 270(2):519–544, 2007

    2007

  15. [23]

    Koivusalo

    H. Koivusalo. Dimension of uniformly random self-similar fractals. Real Anal. Exchange, 39(1):73–90, 2013/14

    2013

  16. [24]

    P. Mattila. Geometry of sets and measures in Euclidean spaces , volume 44 of Cambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 1995. Fractals and rectifiability

    1995

  17. [25]

    P. Mattila. Fourier analysis and Hausdorff dimension , volume 150 of Cambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2015

    2015

  18. [26]

    R. D. Mauldin and S. C. Williams. Random recursive constructions: asymptotic geometric and topological properties. Trans. Amer. Math. Soc., 295(1):325–346, 1986

    1986

  19. [27]

    McMullen

    C. McMullen. The Hausdorff dimension of general Sierpi´ nski carpets.Nagoya Math. J. , 96:1–9, 1984

    1984

  20. [28]

    I. D. Morris and C. Sert. A variational principle relating self-affine measures to self-affine sets. Preprint, available at arXiv:2303.03437, 2023

    2023 arXiv

  21. [29]

    Peres, K

    Y. Peres, K. Simon, and B. Solomyak. Absolute continuity for random iterated function systems with overlaps. J. London Math. Soc. (2) , 74(3):739–756, 2006

    2006

  22. [30]

    Rapaport

    A. Rapaport. On self-affine measures associated to strongly irreducible and proximal systems. Adv. Math., 449:Paper No. 109734, 116, 2024

    2024

  23. [31]

    Solomyak

    B. Solomyak. Measure and dimension for some fractal families. Math. Proc. Cambridge Philos. Soc. , 124(3):531–546, 1998

    1998

  24. [32]

    T. Tao. An introduction to measure theory, volume 126 of Graduate Studies in Mathematics. American Mathematical Society, Providence, RI, 2011. (Bal´ azs B´ ar´ any)Department of Stochastics, HUN-REN–BME Stochastics Research Group, Institute of Mathematics, Budapest University ...

    2011

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.