pith. sign in

arxiv: 2606.05830 · v1 · pith:AGSTEPHLnew · submitted 2026-06-04 · 🧮 math.PR

Biased Random Walk on mathbb Z_+ with Traps of Linearly Increasing Depth

Hua-Ming Wang , Ning Wang This is my paper

Pith reviewed 2026-06-28 00:18 UTC · model grok-4.3

classification 🧮 math.PR
keywords biased random walktrapssub-ballistictransient regimecutpointsdeterministic treelogarithmic growth
0
0 comments X

The pith

Biased random walk on a tree with traps of depth equal to backbone index advances logarithmically when transient.

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

The paper analyzes a λ-biased random walk on a deterministic tree whose backbone vertices each carry an attached trap of length exactly equal to the vertex index. For bias parameter λ at least 1 the walk returns to the root infinitely often, while for λ between 0 and 1 it escapes to infinity but only at logarithmic speed. The precise growth rates are given by explicit constants involving log(1/λ): the liminf of distance over log n equals 1 over that log, and the limsup equals twice that value, almost surely. A further result separates spatial regeneration, which occurs at positive linear density along the backbone, from temporal regeneration, whose count grows only logarithmically with time.

Core claim

When 0 < λ < 1 the walk is transient and its distance |X_n| from the root satisfies liminf |X_n| / log n = 1 / log(1/λ) and limsup |X_n| / log n = 2 / log(1/λ) almost surely. Cutpoints among the first n backbone sites have asymptotic density 1 - λ, while the number of cut times up to time N grows asymptotically as (1 - λ) log N / log(1/λ).

What carries the argument

The deterministic traps of exact depth i attached at each backbone vertex i, which produce sojourns whose lengths grow with i and dominate the long-term displacement.

If this is right

  • Cutpoints occur with positive linear density 1 - λ along the backbone.
  • Cut times up to N grow only logarithmically: M(N) ~ (1 - λ) log N / log(1/λ).
  • The walk is recurrent for every λ ≥ 1.
  • Spatial density of regeneration points stays positive while the corresponding time gaps remain logarithmic.

Where Pith is reading between the lines

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

  • Models with trap depth growing slower or faster than linear in position would likely produce different growth exponents.
  • The separation between linear spatial density and logarithmic temporal density may appear in other biased walks on graphs with increasing trap sizes.
  • Explicit constants of this form could serve as benchmarks for numerical simulations of similar trap models.

Load-bearing premise

The trap depths are exactly equal to the backbone index so that the induced holding times produce the claimed logarithmic scaling.

What would settle it

Run many independent realizations up to large n and check whether |X_n| / log n remains sandwiched between 1/log(1/λ) and 2/log(1/λ) with probability approaching 1.

read the original abstract

We study a $\lambda$-biased random walk $(X_n)_{n\ge0}$ on the deterministic infinite rooted tree $\mathcal{T}=\{(i,j): i\ge0,\,0\le j\le i\}$, whose backbone is $\{(i,0):i\ge0\}$ and, for each $i\ge1$, the segment $\{(i,j):1\le j\le i\}$ forms a trap attached to $(i,0)$. The trapping effect induces long sojourns, yielding asymptotics markedly different from simple random walks. The walk is recurrent for $\lambda\ge1$ and transient for $0<\lambda<1$. In the transient regime it is sub-ballistic: its distance from the root grows logarithmically, with \[ \liminf_{n\to\infty}\frac{|X_n|}{\log n}=\frac{1}{\log(1/\lambda)},\quad \limsup_{n\to\infty}\frac{|X_n|}{\log n}=\frac{2}{\log(1/\lambda)},\quad\text{a.s.}. \] A contrast between spatial and temporal regeneration emerges. Let $C(n)$ be the number of cutpoints among the first $n$ backbone vertices and $M(N)$ the number of cut times up to time $N$. Then \[ \lim_{n\to\infty}\frac{C(n)}{n}= 1-\lambda,\qquad \lim_{N\to\infty}\frac{M(N)}{\log N}=\frac{1-\lambda}{\log(1/\lambda)},\quad\text{a.s.}, \] so cutpoints have positive linear density while cut times grow only logarithmically.

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 / 1 minor

Summary. The manuscript studies a λ-biased random walk on the deterministic rooted tree T = {(i,j) : i ≥ 0, 0 ≤ j ≤ i} with traps of depth exactly i attached to each backbone vertex (i,0). It establishes recurrence for λ ≥ 1 and transience for 0 < λ < 1. In the transient regime the position satisfies liminf |X_n|/log n = 1/log(1/λ) and limsup |X_n|/log n = 2/log(1/λ) almost surely. It further proves that the proportion of cutpoints among the first n backbone vertices converges to 1-λ a.s., while the number of cut times up to time N satisfies M(N)/log N → (1-λ)/log(1/λ) a.s.

Significance. The explicit almost-sure constants for the logarithmic growth, obtained via renewal decomposition along the backbone and geometric scaling of mean sojourn times (1/λ)^i, constitute a precise description of how linearly increasing trap depths dominate the asymptotics. The contrast between linear spatial density of cutpoints and logarithmic growth of cut times is a notable contribution to the analysis of biased walks in inhomogeneous environments.

minor comments (1)
  1. [Introduction / Model] The model definition in the introduction could explicitly state the transition probabilities at backbone vertices to confirm that the bias λ governs the probability of moving toward the root versus entering a trap.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for the positive recommendation to accept. No major comments were raised in the report.

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The paper defines the λ-biased walk on the explicit deterministic tree with traps of depth exactly i at backbone site i. The claimed liminf/limsup constants 1/log(1/λ) and 2/log(1/λ) are obtained by direct renewal analysis of mean sojourn times, which scale geometrically as (1/λ)^i by the bias definition itself; the cutpoint density 1-λ likewise follows from the same return probability at each backbone vertex. No parameter is fitted to data and then relabeled a prediction, no self-citation supplies a uniqueness theorem, and no ansatz is imported. The statements are therefore independent consequences of the model definition rather than reductions to their own inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claims rest on the deterministic tree construction with index-proportional trap depths and on the standard definition of λ-bias; no new entities are introduced and no parameters are fitted to data.

free parameters (1)
  • λ
    Bias parameter varied continuously; determines the recurrence threshold and the explicit speed constants.
axioms (2)
  • standard math Standard properties of biased random walks on graphs (transition probabilities proportional to λ toward the parent).
    The bias mechanism is taken from prior literature on biased walks.
  • domain assumption The infinite deterministic tree T with backbone {(i,0)} and traps of depth exactly i at each i.
    This is the model geometry on which recurrence, transience, and all asymptotic statements depend.

pith-pipeline@v0.9.1-grok · 5827 in / 1442 out tokens · 47701 ms · 2026-06-28T00:18:21.743914+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

23 extracted references

  1. [1]

    Benjamini, O

    I. Benjamini, O. Gurel-Gurevich, O. Schramm. Cutpoints and resistance of random walk paths.Ann. Probab.39 (3): 1122-1136, 2011

    2011

  2. [2]

    Cs ´aki, A

    E. Cs ´aki, A. F¨oldes, P. R´ev´esz. On the number of cutpoints of the transient nearest neighbor random walk on the line.J. Theoret. Probab.23 (2): 624-638, 2010

    2010

  3. [3]

    Denisov, D., Korshunov, V

    D. Denisov, D., Korshunov, V . Wachtel.Markov Chains with Asymptotically Zero Drift: Lamperti’s Problem.Cambridge University Press, 2025

    2025

  4. [4]

    Diaconis, D

    P. Diaconis, D. Freedman. De Finetti’s theorem for Markov chains.Ann. Probab.8: 115- 130, 1980. 20

    1980

  5. [5]

    Durrett.Probability: Theory and Examples(5th ed.)

    R. Durrett.Probability: Theory and Examples(5th ed.). Cambridge University Press, 2019

    2019

  6. [6]

    Erd ˝os, S

    P. Erd ˝os, S. J. Taylor. Some intersection properties of random walk paths.Acta Math. Acad. Sci. Hungar .11: 231-248, 1960

    1960

  7. [7]

    Halberstam, T

    N. Halberstam, T. Hutchcroft. Most transient random walks have infinitely many cut times. Ann. Probab.51 (5): 1932-1962, 2023

    1932

  8. [8]

    T. E. Harris. First passage and recurrence distributions.Trans. Amer . Math. Soc.73, 471- 486, 1952

    1952

  9. [9]

    James, R

    N. James, R. Lyons, Y . Peres. A transient Markov chain with finitely many cutpoints. In Probability and Statistics: Essays in Honor of David A. Freedman, IMS Collections 2: 24-29, 2008

    2008

  10. [10]

    James, Y

    N. James, Y . Peres. Cutpoints and exchangeable events for random walks.Teor . V eroyat- nost. i Primenen.41 (4): 854-868, 1996; English translation inTheory Probab. Appl.41 (4): 666-677, 1997

    1996

  11. [11]

    Kesten, M

    H. Kesten, M. V . Kozlov, F. Spitzer. A limit law for random walk in a random environment. Compos. Math.30(2): 145-168, 1975

    1975

  12. [12]

    Lamperti

    J. Lamperti. Criteria for the recurrence or transience of stochastic processes I.J. Math. Anal. Appl.1 (3-4): 314-330, 1960

    1960

  13. [13]

    Lamperti, A new class of probability limit theorems,J

    J. Lamperti, A new class of probability limit theorems,J. Math. Mech.11(5): 749-772, 1962

    1962

  14. [14]

    G. F. Lawler. Escape probabilities for slowly recurrent sets.Probab. Theory Related Fields 94 (1): 91-117, 1992

    1992

  15. [15]

    G. F. Lawler. Cut times for simple random walk.Electron. J. Probab.1: no. 13, 1-24, 1996

    1996

  16. [16]

    C. H. Lo, M. V . Menshikov, A. R. Wade. Cutpoints of non-homogeneous random walks. ALEA Lat. Am. J. Probab. Math. Stat.19(1): 493-510, 2022

    2022

  17. [17]

    Menshikov, S

    M. Menshikov, S. Popov, A. Wade.Non-homogeneous Random Walks: Lyapunov Func- tion Methods for Near-Critical Stochastic Systems.Cambridge Tracts in Math. 209, Cam- bridge University Press, Cambridge, 2017

    2017

  18. [18]

    A. N. Shiryaev.Probability. 2nd ed. Springer-Verlag, Berlin, Heidelberg, New York, 1996

    1996

  19. [19]

    F. Solomon. Random walks in a random environment.Ann. Probab.3 (1): 1-31, 1975

    1975

  20. [20]

    A. S. Sznitman.Topics in random walks in random environment.School and Conference on Probability Theory, ICTP Lecture Notes Series, Trieste, 2004, 17: 203-266

    2004

  21. [21]

    M. V oit. Strong laws of large numbers for random walks associated with a class of one- dimensional convolution structures.Monatsh. Math.113: 59-74, 1992

    1992

  22. [22]

    H.-M. Wang, L. Wang. Local time, upcrossing time and weak cutpoints of a spatially inhomogeneous random walk on the line.Stochastic Process. Appl.2025, 181: Paper No. 104550, 1-29

    2025

  23. [23]

    Zeitouni,Random walks in random environment.Lecture Notes in Math

    O. Zeitouni,Random walks in random environment.Lecture Notes in Math. 1837, J. Picard (Ed.), 2004, 189-312, Springer-Verlag Berlin Heidelberg. 21

    2004