Pith. sign in

REVIEW 6 minor 39 references

Pointwise subexponential growth and near-diffusive displacement on bounded-degree graphs with non-negative Ollivier--Ricci curvature

T0 review · 0 major / 6 minor · reviewed 2026-07-30 · grok-4.5

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

Pith's one-line read Bounded-degree graphs with non-negative Ollivier–Ricci curvature have pointwise near-diffusive walks and subexponential volume growth at every root.

desk verdict Pointwise upgrade of Hutchcroft–Münch under Ollivier curvature: clean LH bootstrap plus a working discrete Wang coupling; solid and worth engaging. read the letter →

arxiv 2607.27162 v1 pith:ZPWX5BWL submitted 2026-07-29 math.DG math.COmath.PR

Chiyu Zhou This is my paper · ORCID
classification math.DGmath.COmath.PR MSC 53C2160J2705C8158J65
keywords Ollivier-Riccicurvaturerandomwalksongraphsvolumegrowthlog-Harnackinequalitydisplacementestimatesbounded-degreeheatentropy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper shows that a purely local curvature condition on a graph already controls global geometry and random-walk speed, uniformly from every starting vertex. If degrees are bounded and Ollivier–Ricci curvature is non-negative, continuous-time random-walk displacement is at most t times a slowly growing subexponential factor, and the logarithm of ball volume grows at most like exp of the square root of log r times log log r. Earlier results gave only averaged or first-moment statements; the advance is a pointwise bound that holds for every root on possibly infinite graphs. The argument works by turning curvature into a log-Harnack inequality and then running a self-improving loop among displacement, heat entropy, and volume. A sympathetic reader cares because this is a discrete Bishop–Gromov-type statement: local non-negative curvature still forces almost-diffusive large-scale behavior.

What carries the argument

The log-Harnack inequality LH(A): relative entropy between heat kernels started at neighboring points is at most A times (dist²/t + dist/√t). Non-negative Ollivier curvature supplies LH(32d) by a coupling-by-change-of-measure that adds a geodesic-drift jump. LH(A) then feeds a bootstrap loop—displacement controls entropy, entropy controls volume, volume improves displacement—whose iteration yields the subexponential bounds.

What would settle it

Exhibit a single infinite connected graph of bounded degree and non-negative Ollivier–Ricci curvature whose ball volumes satisfy log Vol(B(x,r)) / exp(c √(log r log log r)) → ∞ for every c, or whose continuous-time walk satisfies E_x dist(x,X_t)^2 / (t exp(c √(log t log log t))) → ∞ for every c.

Watch

Extended reading notes

Core claim

On any connected graph of maximum degree d with non-negative Ollivier–Ricci curvature there is a constant C_d such that, for every vertex x and all t,r at least e^e, the continuous-time walk satisfies E_x dist(x,X_t)^2 ≤ t exp[C_d √(log t log log t)] and log Vol(B(x,r)) ≤ exp[C_d √(log r log log r)]. The same near-diffusive bound holds for the discrete-time lazy walk. The estimates are pointwise in the root, not merely averaged.

Load-bearing premise

That non-negative curvature always lets you couple two lazy walks so the second particle can drift one step closer along a geodesic at a uniformly positive rate, without changing the law of the first particle.

Editorial extensions

If this is right

  • Pointwise subexponential volume growth holds on every (possibly infinite) bounded-degree non-negatively curved graph, not only on finite or unimodular ones.
  • The same graphs have pointwise near-diffusive continuous- and discrete-time random walks from every root.
  • Under the extra hypothesis of truly diffusive moments, the log-Harnack inequality upgrades to genuine volume doubling.
  • Any future proof of the Hutchcroft–Münch polynomial-growth conjecture can start from these pointwise subexponential bounds rather than from averaged estimates.

Reading between the lines

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

  • The remaining gap to the conjectured polynomial volume bound is now purely quantitative: one must remove the iterated-log factors produced by the bootstrap, not invent a new averaging device.
  • The same LH-plus-bootstrap loop should apply verbatim to other discrete curvatures (Forman, Bakry–Émery) once a comparable log-Harnack inequality is available.
  • On transitive or unimodular examples the pointwise bounds immediately recover and slightly strengthen the earlier averaged theorems without ergodic decomposition.
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 / 6 minor

Summary. The paper proves that every (possibly infinite) connected graph of bounded degree and non-negative Ollivier–Ricci curvature satisfies pointwise near-diffusive displacement and pointwise subexponential volume growth: uniformly in the root x, E_x dist(x,X_t)^2 ≤ t exp[C_d √(log t log log t)] and log Vol(B(x,r)) ≤ exp[C_d √(log r log log r)] for t,r ≥ e^e. The argument is organized in three blocks: (i) non-negative Ollivier curvature plus deg ≤ d implies a log-Harnack inequality LH(32d) via a geodesic-drift change-of-measure coupling; (ii) LH(A) alone yields the stated bounds by a self-improving displacement–entropy–volume bootstrap whose iteration depth is optimized to √(log t / log log t); (iii) LH(A) plus a diffusive-moment hypothesis implies volume doubling. Discrete-time displacement is recovered from the continuous-time statement by a Poissonization comparison.

Significance. The result upgrades the averaged subexponential bounds of Hutchcroft–Münch to uniform pointwise control on every rooted ball and every starting point, including infinite graphs. This is a genuine local-to-global advance under Ollivier curvature and supplies the strongest available evidence toward their polynomial-growth/diffusive conjecture. The modular architecture (curvature ⇒ LH; LH ⇒ bootstrap; LH + diffusion ⇒ doubling) is clean, the constants are tracked explicitly through the loop, and the coupling argument is a carefully executed discrete analogue of Wang’s change-of-measure method. These features make the paper a substantial and reusable contribution to discrete curvature and geometric analysis on graphs.

minor comments (6)
  1. [§3, Heat entropy] In §3 (Heat entropy), the same symbol H_t(x) is used both for the m-weighted heat entropy and for Shannon entropy; the subsequent identity H_t = H_t + E log m(X_t) is therefore ambiguous in plain text. Introduce a distinct notation (e.g., script or tilde) for Shannon entropy and keep it consistent through Lemmas 3.3–3.5.
  2. [Lemma 3.6] Lemma 3.6: the factor 4/η in the conclusion is slightly loose relative to the integral lower bound η/(2 M^{-a}); a one-line remark that any constant >2/η works would help the reader track the later C_E(A).
  3. [Theorem 3.11] Theorem 3.11: the universal c_0 appearing in the Davies–Gaffney–Grigor’yan lower bound (3.10) is never given a numerical value. Since C^* := 2^{5/3} c_0^{-1} log 2 enters the stage constants, either cite a concrete c_0 from the literature or note that any positive c_0 is absorbed into C_D(A).
  4. [§5, Lemma 5.3] Proposition 5.4 / Lemma 5.3: the construction that forces π_{a,b}(a,b_-)=K(b,b_-) while preserving the W_1 bound is correct, but the verification that the modified plan remains non-negative relies on π(a,b)≥1/2. A short parenthetical recalling K(a,a)≥3/4 and K(b,V\{b})≤1/4 would make the argument self-contained for readers unfamiliar with the lazy-kernel arithmetic.
  5. [Throughout] Several minor typographical inconsistencies appear: “locally-finite” vs “locally finite”, “Ollivier-Ricci” vs “Ollivier–Ricci”, and the future date “July 2026” on the title page. Normalize hyphenation and fix the date before publication.
  6. [Theorem 3.12] In the display after (3.19), the recursive inequality for log(1+M_{n+1}) is written with an implicit absorption of lower-order terms into C(A); stating the precise inductive hypothesis used for the bound log(1+M_n)≤C(n+2)log(e+n) would remove any doubt about the constant bookkeeping.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation is a self-contained bootstrap from Ollivier curvature to LH(A) to subexponential bounds.

full rationale

The paper’s central claim (Theorem 2.1) is obtained by two independent, fully written arguments: (i) non-negative Ollivier–Ricci curvature plus bounded degree implies the log-Harnack inequality LH(32d) via an explicit geodesic-drift coupling and Girsanov change-of-measure (Section 5, Theorem 5.1 / 2.6); (ii) any Markov kernel satisfying LH(A) yields the stated pointwise near-diffusive displacement and subexponential volume growth by a self-improving displacement–entropy–volume loop whose stage constants are tracked and whose iteration count is optimized to √(log t log log t) (Section 3, Theorems 3.7–3.12 and 2.4). Intermediate objects (heat entropy H_t, LH(A)) are defined from first principles and then bounded; they are not fitted to the target quantities. Citations to Hutchcroft–Münch supply only the prior averaged results that the paper strictly strengthens, and are not used as load-bearing lemmas for the pointwise claim. Wang’s coupling method is cited as methodological inspiration, but the discrete construction (Lemmas 5.2–5.3, Proposition 5.4, entropy identities (5.4)–(5.11)) is carried out in full inside the manuscript. There is no data fitting, no uniqueness theorem imported from the same authors, and no renaming of a known empirical pattern. The derivation is therefore self-contained against its stated assumptions.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

The paper works entirely inside standard Markov-chain and optimal-transport language. The only extra structure is the intermediate log-Harnack condition LH(A), introduced as a convenient sufficient hypothesis, and the classical Davies–Gaffney–Grigor’yan heat-kernel upper bound, which is cited rather than re-proved. No numerical parameters are fitted; C_d and C_A are existentially quantified constants produced by the bootstrap.

assumptions (5)
  • domain assumption Definition of non-negative Ollivier–Ricci curvature: W_1(ˆP(x,·), ˆP(y,·)) ≤ 1 whenever x ∼ y (and the induced path-metric contraction).
    Load-bearing geometric hypothesis; stated in §1 and used throughout §5.
  • standard math Davies–Gaffney–Grigor’yan on-diagonal heat-kernel bound h_t^x(y) ≤ (m(x)m(y))^{-1/2} exp(-ζ_t(dist(x,y))) with the elementary lower estimate (3.10) on ζ.
    Invoked as Lemma 3.10 from Bauer–Hua–Yau; controls the volume-to-displacement step (Theorem 3.11).
  • standard math Girsanov change-of-measure formula for continuous-time pure-jump processes on a countable space (entropy cost Ψ_κ(λ)).
    Used in §5 to compute the relative entropy of the controlled coupling versus the reference coupling.
  • domain assumption Degrees bounded by a finite d (equivalently P_min ≥ 1/d > 0).
    Needed both for the constant C_d and to turn the abstract LH(A) into a uniform statement; appears in the main theorem hypotheses.
  • domain assumption Weak reversibility of the Markov kernel (P(x,y)>0 ⇔ P(y,x)>0) so that graph distance is well-defined and symmetric.
    Stated at the opening of §1 and used to define dist and the undirected graph.
invented entities (1)
  • Log-Harnack condition LH(A) independent evidence
    purpose: Abstract sufficient hypothesis that isolates the analytic input needed for the displacement–entropy–volume bootstrap, separating it from the geometric curvature assumption.
    Defined in Definition 2.3; the whole of §3 is written under LH(A) alone. It is a convenient intermediate property rather than a new physical or geometric object, and is later verified for Ollivier-non-negative graphs.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Pointwise subexponential growth and near-diffusive displacement on bounded-degree graphs with non-negative Ollivier--Ricci curvature." pith.science (2026). https://pith.science/paper/ZPWX5BWL

@misc{pith2026260727162,
  author       = {Pith},
  title        = {Pith review of: Pointwise subexponential growth and near-diffusive displacement on bounded-degree graphs with non-negative Ollivier--Ricci curvature},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZPWX5BWL}},
  note         = {Machine review of arXiv:2607.27162}
}
abstract

Let $G=(V,E)$ be a possibly infinite, locally finite graph with non-negative Ollivier--Ricci curvature and degrees bounded by $d<\infty$. We prove that there exists a constant $C_d$ such that the continuous-time random walk displacement and log-volume growth satisfy \[ \mathbb{E}_x \mathrm{dist}(x,X_t)^2 \le t \exp\left[C_d \sqrt{\log t \log\log t}\right], \] \[ \log \mathrm{Vol}(B(x,r)) \le \exp\left[C_d \sqrt{\log r \log\log r}\right], \] for every $x\in V$ and all $r,t \ge e^e$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references

  1. [1]

    Harnack inequality and heat kernel estimates on manifolds with curvature unbounded below.Bull

    Marc Arnaudon, Anton Thalmaier, and Feng-Yu Wang. Harnack inequality and heat kernel estimates on manifolds with curvature unbounded below.Bull. Sci. Math., 130(3):223–233, 2006

    2006

  2. [2]

    On Harnack inequalities and optimal transportation.Ann

    Dominique Bakry, Ivan Gentil, and Michel Ledoux. On Harnack inequalities and optimal transportation.Ann. Sc. Norm. Super. Pisa Cl. Sci. (5), 14(3):705–727, 2015

    2015

  3. [3]

    Ollivier-ricci curvature for causal sets, 2026

    Joe Barton, Samuël Borza, and Jona Röhrig. Ollivier-ricci curvature for causal sets, 2026

    2026

  4. [4]

    Davies-Gaffney-Grigor’yan lemma on graphs

    Frank Bauer, Bobo Hua, and Shing-Tung Yau. Davies-Gaffney-Grigor’yan lemma on graphs. Comm. Anal. Geom., 23(5):1031–1068, 2015

    2015

  5. [5]

    Sharp Davies-Gaffney-Grigor’yan lemma on graphs.Math

    Frank Bauer, Bobo Hua, and Shing-Tung Yau. Sharp Davies-Gaffney-Grigor’yan lemma on graphs.Math. Ann., 368(3-4):1429–1437, 2017

    2017

  6. [6]

    Cambridge University Press, Cambridge, 2021

    TomasBjörk.Point processes and jump diffusions—an introduction with finance applications. Cambridge University Press, Cambridge, 2021

    2021

  7. [7]

    Path coupling without contraction.J

    Magnus Bordewich and Martin Dyer. Path coupling without contraction.J. Discrete Algorithms, 5(2):280–292, 2007

    2007

  8. [8]

    D. P. Bourne, D. Cushing, S. Liu, F. Münch, and N. Peyerimhoff. Ollivier-Ricci idleness functions of graphs.SIAM J. Discrete Math., 32(2):1408–1424, 2018

    2018

Show all 39 references
  1. [9]

    Cover and Joy A

    Thomas M. Cover and Joy A. Thomas.Elements of information theory. Wiley Series in Telecommunications. John Wiley & Sons, Inc., New York, 1991. A Wiley-Interscience Publication

    1991

  2. [10]

    E. B. Davies. Large deviations for heat kernels on graphs.J. London Math. Soc. (2), 47(1):65–72, 1993

    1993

  3. [11]

    Parabolic Harnack inequality and estimates of Markov chains on graphs

    Thierry Delmotte. Parabolic Harnack inequality and estimates of Markov chains on graphs. Rev. Mat. Iberoamericana, 15(1):181–232, 1999

    1999

  4. [12]

    Ricci curvature of finite Markov chains via convexity of the entropy.Arch

    Matthias Erbar and Jan Maas. Ricci curvature of finite Markov chains via convexity of the entropy.Arch. Ration. Mech. Anal., 206(3):997–1038, 2012

    2012

  5. [13]

    Bochner’s method for cell complexes and combinatorial Ricci curvature

    Robin Forman. Bochner’s method for cell complexes and combinatorial Ricci curvature. Discrete Comput. Geom., 29(3):323–374, 2003

    2003

  6. [14]

    Hsu.Stochastic analysis on manifolds, volume 38 ofGraduate Studies in Mathe- matics

    Elton P. Hsu.Stochastic analysis on manifolds, volume 38 ofGraduate Studies in Mathe- matics. American Mathematical Society, Providence, RI, 2002

    2002

  7. [15]

    Tom Hutchcroft and Isaac M. Lopez. A relation between isoperimetry and total variation decay with applications to graphs of non-negative ollivier-ricci curvature, 2024

    2024

  8. [16]

    Bounded-degree graphs of non-negative ollivier-ricci curvature have subexponential growth and diffusive random walk, 2025

    Tom Hutchcroft and Florentin Münch. Bounded-degree graphs of non-negative ollivier-ricci curvature have subexponential growth and diffusive random walk, 2025

    2025

  9. [17]

    On the mean square displacement of a random walk on a graph.European J

    Seonghyuk Im, Hwidong Kim, Jiho Maeng, Jihwan Yu, Yongwook Cha, and Seong-Hun Paeng. On the mean square displacement of a random walk on a graph.European J. Combin., 51:227–235, 2016

    2016

  10. [18]

    Shiryaev.Limit theorems for stochastic processes, volume 288 of Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences]

    Jean Jacod and Albert N. Shiryaev.Limit theorems for stochastic processes, volume 288 of Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences]. Springer-Verlag, Berlin, second edition, 2003. 22

    2003

  11. [19]

    Universitext

    Jürgen Jost.Riemannian geometry and geometric analysis. Universitext. Springer-Verlag, Berlin, 1995

    1995

  12. [20]

    Characterizations of forman curvature, 2021

    Jürgen Jost and Florentin Münch. Characterizations of forman curvature, 2021

    2021

  13. [21]

    Curvature, concentration and error estimates for Markov chain Monte Carlo.Ann

    Aldéric Joulin and Yann Ollivier. Curvature, concentration and error estimates for Markov chain Monte Carlo.Ann. Probab., 38(6):2418–2442, 2010

    2010

  14. [22]

    Klebaner.Introduction to stochastic calculus with applications

    Fima C. Klebaner.Introduction to stochastic calculus with applications. Imperial College Press, London, second edition, 2005

    2005

  15. [23]

    Kullback and R

    S. Kullback and R. A. Leibler. On information and sufficiency.Ann. Math. Statistics, 22:79–86, 1951

    1951

  16. [24]

    Ricci curvature of graphs.Tohoku Math

    Yong Lin, Linyuan Lu, and Shing-Tung Yau. Ricci curvature of graphs.Tohoku Math. J. (2), 63(4):605–627, 2011

    2011

  17. [25]

    Ricci curvature and eigenvalue estimate on locally finite graphs.Math

    Yong Lin and Shing-Tung Yau. Ricci curvature and eigenvalue estimate on locally finite graphs.Math. Res. Lett., 17(2):343–356, 2010

    2010

  18. [26]

    Non-negative Ollivier curvature on graphs, reverse Poincaré inequality, Buser inequality, Liouville property, Harnack inequality and eigenvalue estimates.J

    Florentin Münch. Non-negative Ollivier curvature on graphs, reverse Poincaré inequality, Buser inequality, Liouville property, Harnack inequality and eigenvalue estimates.J. Math. Pures Appl. (9), 170:231–257, 2023

    2023

  19. [27]

    Mixing time and expansion of non-negatively curved Markov chains.J

    Florentin Münch and Justin Salez. Mixing time and expansion of non-negatively curved Markov chains.J. Éc. polytech. Math., 10:575–590, 2023

    2023

  20. [28]

    Wojciechowski

    Florentin Münch and Radosław K. Wojciechowski. Ollivier Ricci curvature for general graph Laplacians: heat equation, Laplacian comparison, non-explosion and diameter bounds.Adv. Math., 356:106759, 45, 2019

    2019

  21. [29]

    Springer, Cham, 2017

    Laurent Najman and Pascal Romon, editors.Modern approaches to discrete curvature, volume 2184 ofLecture Notes in Mathematics. Springer, Cham, 2017

    2017

  22. [30]

    Ricci curvature of Markov chains on metric spaces.J

    Yann Ollivier. Ricci curvature of Markov chains on metric spaces.J. Funct. Anal., 256(3):810–864, 2009

    2009

  23. [31]

    A survey of Ricci curvature for metric spaces and Markov chains

    Yann Ollivier. A survey of Ricci curvature for metric spaces and Markov chains. In Probabilistic approach to geometry, volume 57 ofAdv. Stud. Pure Math., pages 343–381. Math. Soc. Japan, Tokyo, 2010

    2010

  24. [32]

    M. M. H. Pang. Heat kernels of graphs.J. London Math. Soc. (2), 47(1):50–64, 1993

    1993

  25. [33]

    Log-Harnack inequality for stochastic differential equations in Hilbert spaces and its consequences.Infin

    Michael Röckner and Feng-Yu Wang. Log-Harnack inequality for stochastic differential equations in Hilbert spaces and its consequences.Infin. Dimens. Anal. Quantum Probab. Relat. Top., 13(1):27–37, 2010

    2010

  26. [34]

    Sparse expanders have negative curvature.Geom

    Justin Salez. Sparse expanders have negative curvature.Geom. Funct. Anal., 32(6):1486– 1513, 2022

    2022

  27. [35]

    Modern aspects of markov chains: entropy, curvature and the cutoff phe- nomenon, 2025

    Justin Salez. Modern aspects of markov chains: entropy, curvature and the cutoff phe- nomenon, 2025

    2025

  28. [36]

    Curvature of nonlocal Markov generators

    Michael Schmuckenschläger. Curvature of nonlocal Markov generators. InConvex geometric analysis (Berkeley, CA, 1996), volume 34 ofMath. Sci. Res. Inst. Publ., pages 189–197. Cambridge Univ. Press, Cambridge, 1999. 23

    1996

  29. [37]

    Rényi divergence and Kullback-Leibler divergence

    Tim van Erven and Peter Harremoës. Rényi divergence and Kullback-Leibler divergence. IEEE Trans. Inform. Theory, 60(7):3797–3820, 2014

    2014

  30. [38]

    Logarithmic Sobolev inequalities on noncompact Riemannian manifolds

    Feng-Yu Wang. Logarithmic Sobolev inequalities on noncompact Riemannian manifolds. Probab. Theory Related Fields, 109(3):417–424, 1997

    1997

  31. [39]

    Coupling by change of measure, harnack inequality and hypercontractivity

    Feng-Yu Wang. Coupling by change of measure, harnack inequality and hypercontractivity. In Andreas Eberle, Martin Grothaus, Walter Hoh, Moritz Kassmann, Wilhelm Stannat, and Gerald Trutnau, editors,Stochastic Partial Differential Equations and Related Fields, pages 381–389, Ch...

    2018

Pith tools

Reviewed July 30, 2026 · model on record in the stance chip above.