Pith. sign in

REVIEW 6 major objections 5 minor 33 references

Convergence in natural parametrization of random walk frontier

T0 review · 6 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read The random-walk frontier converges to the Brownian frontier in the natural parametrization metric, with time set by 4/3-Minkowski content.

desk verdict The paper claims the long-awaited natural parametrization convergence of the random walk frontier, but several load-bearing estimates are stated without proof, so the central theorem cannot be verified from the text as submitted. read the letter →

arxiv 2504.13743 v1 pith:4R4QFNZZ submitted 2025-04-18 math.PR

Yifan Gao , Xinyi Li , Runsheng Liu , Xiangyi Liu , Daisuke Shiraishi This is my paper · ORCID
classification math.PR MSC 60J6560G5060F1728A80
keywords randomwalkfrontierBrowniannaturalparametrizationMinkowskicontentoccupationmeasuredisconnectionexponentGreen'sfunctionscalinglimit
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 claims that the outer boundary of planar simple random walk, drawn on the lattice $e^{{-n}}$$Z^{2}$ and run until it exits the unit disk, converges weakly as a parametrized curve to the outer boundary of planar Brownian motion. The convergence is in the natural parametrization metric, which compares both the shape of a curve and the way time is assigned along it. The time parameter on the Brownian side is the 4/3-Minkowski content of the frontier, so the result says the random walk's own traversal speed—each edge taking time c1 $e^{{-4n/3}}$—converges to the natural fractal clock of the Brownian frontier. The intermediate claim is that the renormalized occupation measure on random-walk frontier points converges to the 4/3-Minkowski content measure. A sympathetic reader would care because this upgrades the known Hausdorff-metric convergence of frontiers to a genuine curve-level scaling limit, matching the conjectured 4/3-dimensional structure of the Brownian boundary.

What carries the argument

The central objects are the frontier of a lattice path and the frontier-disk events: a small disk is a frontier disk if the path enters it, the two arms around it do not disconnect it from infinity, and the intermediate excursion stays close by. The one-point and two-point frontier Green's functions G_fr_D(z) and G_fr_D(z,w) are the limits of the normalized expected weights of these events, and they supply the first and second moments needed for the L2 estimate. The sharp non-disconnection estimate P_{0,0}(D_m) ≃ q $e^{{-αm}}$ with α=2/3, transferred from Brownian motion to simple random walk, is the quantitative input that fixes the constant c1 and the exponential error. The L2-approximation then compares the discrete occupation measure against the continuum frontier-disk measure, and the 4/3-Minkowski content measure ν serves as the time parameter of the limiting curve.

What would settle it

Compute the ratio P(D_m)$e^{{2m/3}}$ for two independent simple random walks that keep the origin on the frontier; if for large m the ratio wanders by more than $e^{{-um}}$ (for example, decays like $m^{{-c}}$), then (2.12) fails and Theorem 1.2 loses its key input. Alternatively, check the frontier Green's function identity E[L_s(z)] ≃ G_fr_D(z) near |z| = 1/2: a polynomial rather than exponential error would invalidate Theorem 3.2 and the L2 estimate (3.9).

Watch

Extended reading notes

Core claim

Theorem 1.1 asserts that γ_n converges weakly to eγ under the natural parametrization metric ρ, where eγ traces the frontier of Brownian motion run until it first exits the unit disk and is parameterized so that Cont_{4/3}(eγ[0,t]) = t. Theorem 1.2 gives the equivalent measure statement: with ν_n = c1 $e^{{-4n/3}}$ Σ_{x∈Z_n∩fr(λ_n)} δ_x and ν(·) = Cont_{4/3}(·∩eγ), the laws of ν_n converge weakly to ν. The proof first establishes one- and two-point frontier Green's function estimates and sharp frontier-disk probabilities, uses an L2-approximation with a Skorokhod embedding to compare discrete and continuum frontier disks, and then turns the occupation-measure convergence into a natural-parametrization statement by proving tightness and uniqueness of subsequential limits. In the author's framing, the occupation-measure convergence is not an add-on; it is the mechanism that forces the time parametrization of the limiting curve to be the 4/3-Minkowski content measure.

Load-bearing premise

The entire proof rests on several sharp estimates—one-point and two-point frontier Green's functions and the exponential-error non-disconnection estimate for simple random walks—that are stated without full proofs in this paper; if any of them holds only with polynomial error, the occupation-measure convergence and the natural parametrization result collapse.

Editorial extensions

If this is right

  • The natural parametrization of the Brownian frontier is the weak limit of the random walk frontier's own traversal time, so the exponent 4/3 appears as a traversal-time exponent for lattice frontiers.
  • The renormalized occupation measure of frontier points converges to a non-atomic measure, so along the limiting curve no point is visited for a positive amount of time, ruling out 'stuck' subsequential limits.
  • The frontier Green's function estimates give an SLE-free route to the existence of the 4/3-Minkowski content of the Brownian frontier, as stated in Theorem 3.4.
  • Combined with the earlier Hausdorff-metric convergence, the result identifies the Brownian frontier as the unique subsequential limit of random walk frontiers under both the reparametrization metric and the natural parametrization metric.

Reading between the lines

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

  • If the deferred sharp estimates are completed as stated, the same L2 scheme should transfer to other lattice curves whose boundaries admit analogous one- and two-point Green's function estimates, such as killed or conditioned random walks in domains.
  • A direct numerical check of P(D_m)e^{2m/3} for moderate m would give an early indication of whether the exponential error in (2.12) is plausible before a full proof appears.
  • The method suggests that for any scale-invariant planar curve with the same two-arm disconnection exponent, the natural time should be its 4/3-Minkowski content; a testable extension is whether the frontiers of other lattice models obey the same occupation-measure limit.
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

6 major / 5 minor

Summary. The paper claims two main results. Theorem 1.1 states that the frontier of a simple random walk on the scaled lattice Z_n = e^{-n}Z^2, traversed with each edge taking time c_1 e^{-4n/3}, converges weakly under the natural parametrization metric rho to the frontier of planar Brownian motion, with the Brownian frontier parameterized by its 4/3-Minkowski content measure. Theorem 1.2 states that the associated renormalized occupation measure nu_n = c_1 e^{-4n/3} sum_{x in Z_n cap fr(lambda_n)} delta_x converges weakly to the Minkowski content measure nu of the Brownian frontier. The proof strategy is to establish one- and two-point frontier Green's function estimates, prove an L2 approximation between nu_n and nu, use this to derive convergence of occupation measures, and then combine this with convergence modulo reparametrization, tightness under rho, and uniqueness of subsequential limits. The overall architecture follows the cut-point program of the authors' earlier work and uses Skorokhod and KMT couplings, non-disconnection exponents, quasi-invariant measures, and Aizenman-Burchard tightness.

Significance. If the missing estimates are supplied, the paper would be a significant advance: it upgrades the known Hausdorff-metric convergence of random walk frontiers to convergence in the stronger natural-parametrization metric and identifies the normalization constant c_1 in terms of the two-arm disconnection exponent. The paper's strategy is coherent, and the appendices do contain detailed proofs of several substantial propositions, including Propositions 4.2, 4.3, and 4.4 and Lemma D.1; the use of thickenings and quasi-invariant measures is appropriate. However, the central load-bearing statements are not proved in this manuscript: the sharp frontier Green's function estimates, the sharp SRW non-disconnection estimate, the inward coupling theorem, the discrete-continuum event comparison lemma, the two-point moment identification, and the final L2 estimate are either stated without proof or only sketched. As submitted, the main theorems are therefore conditional on a large number of unproved inputs.

major comments (6)
  1. [Section 4.1, Theorem 4.1(ii)] The proof of Theorem 4.1 proves only part (i). Part (ii), the two-point estimate c_1^2 e^{2 alpha n} P(A_n(z) cap A_n(w)) simeq_V G^fr_D(z,w), is never established. There is no two-point analogue of Proposition 4.4 or Proposition 4.5, no discussion of the uniformity of the implied constant in V, and no proof that the cross-terms are exponentially small. This estimate is directly used in the derivation of (5.13) in Section 5.2, which is the engine of Theorem 1.2 and, through it, Theorem 6.7. Without a proof of part (ii), the two-point moment identification underlying convergence of the occupation measure is unsupported.
  2. [Section 2.2, Eq. (2.12)] The sharp SRW non-disconnection estimate P_{0,0}(D_m) simeq q e^{-alpha m} is asserted without proof, with the text saying the proof is 'omitted for brevity' and with the constant q left unspecified. This is load-bearing because the proof of Theorem 4.1(i) defines c_1 = 1/(c_2 q), so the normalization of nu_n in Theorem 1.2 and the parametrization speed in Theorem 1.1 inherit this unproved constant. The exponential error in (2.12) is also used throughout the moment computations. A complete proof or a precise published reference for this sharp estimate is necessary.
  3. [Section 3, Theorems 3.2-3.5] The one- and two-point frontier Green's function estimates, the existence of the non-atomic measure nu with moments given by (3.8), and the L2 estimate (3.9) for the continuum frontier-disk measure are all stated as theorems but their proofs are omitted. The text says the proofs follow 'similarly as in their cut-point counterparts with minor technical modifications', but these are not minor or local results: Theorem 3.3 is used in the two-point estimate and Theorem 3.5 is used directly in the continuum half of (5.13). If any of these statements has only polynomial error instead of the asserted e^{-us} error, the L2 convergence and hence the natural-parametrization identification break. These theorems need either full proofs or a precise statement of the dependence on the cited works [3] and [5].
  4. [Sections 4.2 and Appendix D] Theorem 4.6 (inward coupling for non-disconnecting walks) and Lemma D.2, which is said to imply Proposition 4.5, are both stated without proof. Theorem 4.6 is used essentially in Lemma 4.10 and hence in Propositions 4.8 and 4.9, while Lemma D.2 underlies the comparison in Proposition 4.5 that enters the proof of Theorem 4.1. The statements themselves are nontrivial, and the required exponential error rate e^{-u(n-m)} or simeq cannot simply be assumed from analogous cut-point results. Omitted proofs of these two statements leave the discrete-to-continuum comparison unverified at the precision needed by the paper.
  5. [Section 4.1, proof of Theorem 4.1(i)] The chain of equivalences in (4.11) uses Proposition 4.2 to replace the sum over Y_{n/6}(z_n) by the sum over NICE_{n/6}(z_n). Proposition 4.2 is stated for n >= m >= 10l. In this application l = n/6, so the condition requires m >= 10l = 5n/3, which is incompatible with m <= n. Thus the stated Proposition 4.2 cannot be applied at the scale needed in (4.11). Either Proposition 4.2 needs a different formulation, or the proof of Theorem 4.1(i) needs a different scale choice; as written, the replacement of Y_{n/6} by NICE_{n/6} is unjustified.
  6. [Section 6.2.1, Theorem 6.5] The proof of tightness under rho asserts that nu_n(V) -> nu(V) in probability for any rational disk V and then invokes (5.13). However, (5.13) is proved only for nice boxes V in V. The paper does not prove the extension from nice boxes to arbitrary rational disks, although such an extension may be obtainable by approximating disks by nice boxes and using non-atomicity of nu. As written, this is a gap in the proof of Theorem 6.5 and therefore in the proof of Theorem 1.1.
minor comments (5)
  1. [Eq. (1.3)] The metric rho as defined appears asymmetric in the two curves: the infimum is taken only over bijections alpha : [0,t_gamma] -> [0,t_{gamma'}], with no analogous reparametrization of gamma'. The natural parametrization metric usually involves reparametrizations of both curves; the formula should be corrected or clarified.
  2. [Section 2.2, (2.8) and (2.12)] The constant q is said in (2.8) to depend on the initial configuration beta, while in (2.12) it is called simply 'a constant' for the origin-to-origin case. The paper later uses this q in defining c_1 without discussing whether the same constant applies after translation to z_n; the domain is bounded, so translation invariance is not automatic and the use of a single q should be justified.
  3. [Section 3.1, Definition 3.1] The notation D_{-s}(z) and D_{-2s/3}(z) for disks is nonstandard and is never explicitly defined; the reader must infer that these are disks of radius e^{-s} and e^{-2s/3} respectively.
  4. [Appendix C] In the discussion after (C.1), 'total variance norm' should be 'total variation norm'.
  5. [Section 5.2] The paper says 'we only provide a proof sketch' and omits the approximation of continuous test functions by step functions in the proof of (5.12). This is presented as following 'the same lines' as in [3]; for a self-contained proof, at least the key steps of that approximation should be included.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular step: occupation-measure convergence and the natural-parametrization identification are derived from independent moment estimates, though several key estimates are deferred to self-cited work.

full rationale

Walking the derivation chain: Theorem 1.1 is obtained from Theorem 6.1 (convergence modulo reparametrization) plus convergence of occupation measure, i.e. Theorem 1.2 and estimate (5.13). Theorem 1.2 is in turn proved by a standard L2 moment argument using one-point and two-point frontier Green's function estimates (Theorem 4.1) and frontier-disk controls (Propositions 4.8-4.9, Lemmas 5.2-5.5). The constant c1 is not a fitted prediction: it is defined in Theorem 4.1(i) as 1/(c2 q) after deriving P(A_n(z)) ~ c2 G^fr_D(z) q e^{-alpha n}; it is a normalizing constant fixed by the one-point estimate, not a parameter chosen to force the stated weak limit. No equation in the paper is the target theorem restated as an input, and I could not exhibit any reduction of the form 'Eq. X = Eq. Y by construction' or 'fitted parameter renamed as prediction'. The two-point estimate (4.3) and the sharp non-disconnection estimate (2.12) are independent inputs to (5.13), not consequences of Theorem 1.2 or Theorem 1.1. What is real, and worth flagging explicitly, is a serious support gap rather than circularity: the paper explicitly defers several load-bearing results. Section 3 states that Theorems 3.2-3.3 are 'frontier-point analogues' of [5] with proofs 'omitted'; Theorem 3.5 is omitted by similarity to Theorem 6.14 of [3]; (2.12) is asserted with an unspecified constant q and 'omitted for brevity'; Theorem 4.6 is omitted by similarity to Theorem 8.1 of [3]; and Section 5.2 says (5.13) follows 'following the same lines as in the proof of Proposition 10.1 of [3]' with 'further details' omitted. These are self-citations to the authors' own prior cut-point work [3,5], and they are load-bearing in the sense that the frontier analogues are not proved here. However, the cited works concern cut points, not the frontier conclusion, so the dependence is a hereditary proof obligation, not a circular definition. I therefore score 2: no circular derivation, but a non-negligible self-citation and omitted-proof burden that a fully self-contained version would need to remove.

Assumptions & free parameters 3 free parameters · 11 assumptions · 0 invented entities

The central claim is built on exponents and estimates from prior literature, plus a sheaf of unproved or deferred statements specific to this paper. The scaling exponent 4/3 = 2 - ξ(2) is derived from proven intersection exponents, and the constant c1 is derived from one-point asymptotics rather than fit, but its explicit value depends on constants c2 and q that are asserted without proof. The load-bearing deferred statements are Theorems 3.2-3.5, Theorem 4.6, Lemma D.2, and the final L2 step in Section 5.2.

free parameters (3)
  • c1 = unspecified (defined as 1/(c2 q))
    Time scaling constant in νn and γn; fixed by Theorem 4.1 so that the first moment of the discrete occupation measure matches the continuum Green's function. Its explicit value depends on the unspecified constants c2 and q.
  • c2 = unspecified positive constant
    Introduced in Proposition 4.4's asymptotic equivalence (4.9); existence and positivity are asserted but the proof is deferred.
  • q = unspecified positive constant
    Constant in the sharp SRW non-disconnection estimate (2.12), which is stated without proof. It enters Theorem 4.1 and thereby fixes c1; if (2.12) failed, the scaling constant would shift.
assumptions (11)
  • ad hoc to paper Sharp SRW non-disconnection estimate (2.12): P_{0,0}(D_m) ≃ q e^{-αm} with α = ξ(2) = 2/3, constant q unspecified
    Stated in Section 2.2 with the proof 'omitted for brevity'; used in the proof of Theorem 4.1(i) to evaluate the sum over configurations and set the value of c1.
  • ad hoc to paper One- and two-point frontier Green's function estimates, Theorems 3.2 and 3.3: E[L_s(z)] ≃ G^fr_D(z) and E[L_s(z)L_s(w)] ≃ G^fr_D(z,w)
    Proofs are omitted ('follow similarly as in their cut-point counterparts... therefore omitted'); these estimates feed Theorem 4.1 and the moment computation for occupation measures.
  • ad hoc to paper Theorem 3.4: existence of a non-atomic measure ν with E[ν(V)] = ∫_V G^fr_D(z) dz and E[ν(V)^2] = ∫_V G^fr_D(z,w) dzdw
    Proof omitted; defines the limiting occupation measure and non-atomicity, used in Lemma 6.6 and Theorem 6.7 to identify subsequential limits.
  • ad hoc to paper Theorem 3.5: E[eν_s(V) - ν(V)]^2 = O_V(e^{-us}) for the continuum frontier-disk measure
    The paper says it 'decides to omit' the proof; this estimate is the continuum half of the L2 bound (5.13).
  • ad hoc to paper Theorem 4.6 (inward coupling for non-disconnecting walks): P(λ =_m λ*) ≥ 1 - c e^{-u(n-m)}
    Stated without proof, deferred to 'Theorem 8.1 in [3]'; used in Lemma 4.10 to couple the discrete path measures for frontier disks and frontier points.
  • ad hoc to paper Lemma D.2: discrete and continuum frontier-event probabilities are asymptotically equivalent, P(E_n(ζ)) ≃ P(eE_n(ζ))
    Proof is 'analogous to Proposition 7.9 in [3]' and omitted; this comparison underlies Proposition 4.5, which is needed for Theorem 4.1.
  • standard math Intersection and disconnection exponent values from (2.4): ξ(1)=1/4, ξ(2)=2/3, ξ(2,2)=35/12, etc.
    Proven by Lawler-Schramm-Werner [18-22]; used throughout, especially in Lemma 2.1, (2.7), (6.5).
  • standard math One-arm non-disconnection estimate (Lemma 2.1): P_{x0}(eD_{r,R}) ≍ e^{-(R-r)/4}
    Quoted from [12]; used in numerous path-decomposition estimates in Sections 4-6.
  • standard math Separation lemmas (Lemmas 2.3 and 2.5): well-separated non-disconnecting path pairs occur with probability comparable to the base event
    From Lawler [9]; used to establish lower bounds on masses in Lemmas 4.11, 4.14, and Appendix C.
  • standard math Skorokhod embedding (2.17) and strong approximation (2.18)-(2.19) coupling SRW and Brownian motion
    Quoted from [11], [15], [7]; used in Propositions 4.3, 4.5, Lemmas 5.2, 5.5, and Theorem 6.7.
  • standard math Existence and exponential approximation of quasi-invariant measures Q and Q* (equations (C.1), (C.2))
    Invoked in Appendix C for Proposition 4.4; the construction follows [12] and is not proved in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Convergence in natural parametrization of random walk frontier." pith.science (2026). https://pith.science/paper/4R4QFNZZ

@misc{pith2026250413743,
  author       = {Pith},
  title        = {Pith review of: Convergence in natural parametrization of random walk frontier},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4R4QFNZZ}},
  note         = {Machine review of arXiv:2504.13743}
}
read the original abstract

In this paper, we show that the frontier of planar random walk converges weakly under natural parametrization to that of planar Brownian motion. As an intermediate result, we also show the convergence of the renormalized occupation measure.

Figures

Figures reproduced from arXiv: 2504.13743 by the authors.

Figure 6.1
Figure 6.1. Illustration of backtracking and four arms crossing the annulus. Note that typically [PITH_FULL_IMAGE:figures/full_fig_p022_6_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 26 canonical work pages

  1. [5]

    Holden, G

    N. Holden, G. F. Lawler, X. Li, and X. Sun. Minkowski content of Brownian cut points Ann. Inst. Henri Poincar´ e Probab. Stat., 2022, 58(1): 455–488

    work page 2022

  2. [3]

    Y. Gao, X. Li, P. Panov, D. Shiraishi. Scaling limit of the occupation measure of random walk cut points. Preprint, available at arXiv:2310.09592

    work page Pith review arXiv

  3. [1]

    Aizenman, A

    M. Aizenman, A. Burchard. H¨ older regularity and dimension bounds for random curves. Duke Math. J. , 1999, 99(3): 419–453

    work page 1999

  4. [2]

    Duplantier, K.-H

    B. Duplantier, K.-H. Kwon. Conformal invariance and intersection of random walks. Phys. Rev. Lett., 1988, 61(22): 2514–2517

    work page 1988

  5. [4]

    Garban, G

    C. Garban, G. Pete, and O. Schramm. Pivotal, cluster, and interface measures for critical planar percolation. J. Amer. Math. Soc. , 2013, 26(4): 939–1024

    work page 2013

  6. [6]

    Holden, X

    N. Holden, X. Li, and X. Sun. Natural parametrization of percolation interface and pivotal points. Ann. Inst. Henri Poincar´ e Probab. Stat., 2022, 58(1): 7–25

    work page 2022

  7. [7]

    Kozdron, G

    M. Kozdron, G. F. Lawler. Estimates of random walk exit probabilities and application to loop-erased random walk. Electron. J. Probab., 2005, 10: 1442–1467

    work page 2005

  8. [8]

    Kiefer, P

    R. Kiefer, P. M¨ orters. The Hausdorff Dimension of the Double Points on the Brownian Frontier. J. Theor. Probab., 2010, 23: 605–623

    work page 2010

Show all 33 references
  1. [9]

    G. F. Lawler. The dimension of the frontier of planar Brownian motion. Electron. Commun. Probab., 1996, 1(5): 29–47

    1996

  2. [10]

    G. F. Lawler. Hausdorff dimension of cut points for Brownian motion. Probab. Theory Related Fields, 1996, 108(2): 177–193

    1996

  3. [11]

    G. F. Lawler. Cut times for simple random walk. Electron. J. Probab., 1996, 1(13): 1–24

    1996

  4. [12]

    G. F. Lawler. Strict concavity of the intersection exponent for Brownian motion in two and three dimensions. Math. Phys. Electron. J. , 1998, 4(5): 1–67

    1998

  5. [13]

    G. F. Lawler. Conformally invariant processes in the plane . Mathematical Surveys and Monographs, 114. American Mathematical Soc., 2008

    2008

  6. [14]

    G. F. Lawler. Minkowski content of the intersection of a Schramm-Loewner evolution (SLE) curve with the real line. J. Math. Soc. Japan , 2015, 67(4): 1631–1669

    2015

  7. [15]

    G. F. Lawler, V. Limic. Random Walk: A Modern Introduction. Cambridge University Press, 2010

    2010

  8. [16]

    G. F. Lawler, E. E. Puckette. The intersection exponent for simple random walk. Combin. Probab. Comput., 2000, 9(5): 441–464

    2000

  9. [17]

    G. F. Lawler, M. A. Rezaei. Minkowski content and natural parameterization for the Schramm–Loewner evolution. Ann. Probab., 2015, 43(3): 1082–1120

    2015

  10. [18]

    G. F. Lawler, O. Schramm, and W. Werner. The Dimension of the Planar Brownian Frontier is 4/3. Math. Res. Lett., 2001, 8(4): 401–411

    2001

  11. [19]

    G. F. Lawler, O. Schramm, W. Werner. Values of Brownian intersection exponents, I: Half- plane exponents. Acta Math., 2001, 187(2): 237–273. 32

    2001

  12. [20]

    G. F. Lawler, O. Schramm, W. Werner. Values of Brownian intersection exponents, II: Plane exponents. Acta Math., 2001, 187(2): 275–308

    2001

  13. [21]

    G. F. Lawler, O. Schramm, W. Werner. Values of Brownian intersection exponents III: Two-sided exponents. Ann. Inst. Henri Poincar´ e Probab. Stat., 2002, 38(1): 109–123

    2002

  14. [22]

    G. F. Lawler, O. Schramm, W. Werner. Analyticity of intersection exponents for planar Brownian motion. Acta Math., 2002, 189(2): 179–201

    2002

  15. [23]

    G. F. Lawler, O. Schramm, W. Werner. Sharp estimates for Brownian non-intersection probabilities. In and Out of Equilibrium: Probability with a Physics Flavor. Birkh¨ auser Boston, 2002: 113–131

    2002

  16. [24]

    G. F. Lawler, O. Schramm, W. Werner. Conformal restriction: the chordal case. J. Amer. Math. Soc., 2003, 16(4): 917–955

    2003

  17. [25]

    G. F. Lawler, F. Viklund. Convergence of loop-erased random walk in the natural parametrization. Duke Math. J. , 2016, 170(10): 2289–2370

    2016

  18. [26]

    G. F. Lawler, W. Werner. The Brownian loop soup. Probab. Theory Related Fields , 2004, 128(4): 565–588

    2004

  19. [27]

    X. Li, D. Shiraishi. Convergence of three-dimensional loop-erased random walk in the nat- ural parametrization. Probab. Theory Related Fields, 2025, 191(1): 421–521

    2025

  20. [28]

    Mandelbrot

    B. Mandelbrot. The fractal geometry of nature . Freeman, San Francisco, 1982

    1982

  21. [29]

    M¨ orters, Y

    P. M¨ orters, Y. Peres.Brownian motion. Cambridge University Press, 2010

    2010

  22. [30]

    W. Qian. Generalized disconnection exponents. Probab. Theory Related Fields , 2021, 179: 117–164

    2021

  23. [31]

    Shiraishi

    D. Shiraishi. Two-sided random walks conditioned to have no intersections. Electron. J. Probab., 2012, 17(18): 1–24

    2012

  24. [32]

    van de Brug, F

    T. van de Brug, F. Camia, M. Lis. Random walk loop soups and conformal loop ensembles. Probab. Theory Related Fields, 2016, 166: 553–584

    2016

  25. [33]

    D. Zhan. Boundary Green’s functions and Minkowski content measure of multi-force-point SLEκ(ρ). Stoch. Process. Appl., 2022, 151: 265–306. 33

    2022

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.