Sublinear Morse geodesics and First Passage Percolation

Sagnik Jana, Yulan Qing · 2025

Read Sublinear Morse geodesics and First Passage Percolation by Sagnik Jana, Yulan Qing (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-07859.txt.

Introduction

Sagnik Jana and Yulan Qing prove a result about random paths on graphs in this two thousand twenty-five math paper. Take an endless graph and give each edge a random length. The shortest routes through this warped graph are called geodesics. An earlier result showed that, under certain shape conditions, a two-way endless geodesic exists. That result leaned on a notion called Morse geodesics, which stay stable under detours.

Jana and Qing loosen the shape demand. They only require a geodesic that is Morse in a sublinear sense, a weaker grip. With edges of bounded degree, positive lengths, and finite mean, they prove the endless geodesic still almost surely exists. The setting is called first passage percolation. The work extends the theory of random metrics on graphs.

Catalog record

CREATORSagnik Jana, Yulan Qing
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-07859

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Sublinear Morse geodesics and First Passage Percolation Sagnik Jana, Yulan Qing Abstract . Given an infinite connected graph, a way to randomly perturb its metric is to assign random i.i.d. lengths to the edges of the graph. Assume that the graph is infinite and of bounded degree. Assume also strict positivity and finite expectation of the edge length distribution and existence of a sublinearly Morse bi-infinite geodesic line, we prove that almost surely there exists a bi-infinite geodesic line. This generalizes a previous result of [ BT17 ] regarding Morse geodesics. 1. Introduction First passage percolation is a model of random perturbation of a given metric graph. In this paper, we shall restrict to the simplest model, where random i.i.d lengths are assigned to the edges of a fixed graph, and analyse the existence of infinite geodesic lines. We refer to [ ADH15 ] , [ GK12 ] , [ Kes84 ] for background and references. We recall here how first passage percolation is defined following [ BT17 ] . We consider a connected non-oriented graph X , whose set of vertices and edges are denoted by V and E , respectively. For every function \omega:E\rightarrow(0,\infty) , we equip V with the weighted graph metric d_{\omega} , where each edge e has weight \omega(e) . In other words, for every v_{1},v_{2}\in V,d_{\omega}\left(v_{1},v_{2}\right) is defined as the infimum over all path \gamma=\left(e_{1},\ldots,e_{m}\right) joining v_{1} to v_{2} of |\gamma|_{\omega}:=\sum_{i=1}^{m}\omega\left(e_{i}\right) . We note that the infimum is always realized as a consequence of the bounded degree assumption. The graph metric on X corresponds to the case where \omega=1 , in this case, we denote the metric as d . Our model consists in choosing independently at random the weights \omega(e) according to the probability measure defined on the set of all weight functions \omega . We refer to section 2.3 for formal details. Let X be a simplicial graph. Recall that a path \gamma=\left(e_{1},\ldots,e_{n}\right) between two vertices x,y is a sequence of consecutive edges joining x to y . We write x=\gamma(0),\ldots,\gamma(n)=y to mean the set of vertices such that for all 0\leq i<n,\gamma(i) and \gamma(i+1) are joined by the edge e_{i+1} . For all i<j , we shall also denote by \gamma([i,j]) the sub path \left(e_{i+1},\ldots,e_{j}\right) joining \gamma(i) to \gamma(j) . Let X_{\omega} denotes the graph X with edge lengths \omega(e) for each edge e . A bi-infinite geodesic is a geodesic which is doubly infinite that is, its vertices are indexed by \mathbb{Z} . It was conjectured that under i.i.d lengths of edges, almost surely there are no bi-infinite geodesics in FPP on \mathbb{Z}^{2}. See [ Kes84 ] for more details. For negatively curved spaces, the situation changes dramatically. In [ WW98 ] , Wehr proved that almost surely there is no bi-infinite geodesic which lies entirely in the upper half Euclidean plane. On the other hand, [ BT17 ] established the existence of bi-infinite geodesic line in the affirmative if we assume the existence of a Morse quasi-geodesic line before the percolation. In this case the geometric property of a Morse quasi-geodesic line strongly informs how it behaves after the percolation. We are thus inspired by the proof and show in this paper that even weaker assumptions suffices. That is, we suppose instead there is a geodesic line whose Morse probably weakens as it extends to infinity. This class of quasi-geodesics were constructed recently by [ QRT22 , QRT23 ] and named sublinearly Morse quasi-geodesics. The construction roughly replaces bounded neighborhood with a sublinear neighborhood with respect to the distance to the origin. In this paper, we prove Theorem A . Let X be an infinite connected graph with bounded degree. Assume \mathbb{E}\omega_{e}<\infty and \nu(0)=0 . If X contains a sublinearly Morse bi-infinite quasi-geodesic line, then for almost every \omega , there is a bi-infinite geodesic ray in X_{\omega} . The proof is analogous to that of [ BT17 ] . We first offer a new property of sublinearly Morse geodesic rays and lines, which we refer to as middle recurrence. The name comes from the analogous property used in [ BT17 ] . The property is first exhibited by [ DMS09 ] with an improved proof by [ ADT17 ] . Using the middle recurrence, we derive a pointed superlinear divergence property when then is used in the main proof. In an upcoming paper, we aim to study whether sublinearly Morse directions are preserved under first passage percolation. History For the Cayley graph of \mathbb{Z}^{2} with respect to standard generators, first passage percolation was introduced by Hammersley and Welsh [ HW65 ] . Benjamini, Tessera, and Zeitouni [ BZ12 , BT17 ] studied the question in Gromov hyperbolic spaces and a number of questions have been raised there. In particular, [ BZ12 ] established the tightness of fluctuations of the passage time from the center to the boundary of a large ball, and [ BT17 ] established the almost sure existence of bi-geodesics in hyperbolic spaces. Recently Basu-Mj [ BM22 ] studied the first passage percolation on Gromov hyperbolic group G with boundary equipped with the Patterson-Sullivan measure \nu . They showed that the variance of the first passage time grows linearly with word distance along word geodesic rays in every fixed boundary direction. The sublinearly Morse boundary is a geometric boundary construction that generalize the Gromov boundary. It is shown to be always quasi-isometrically invariant and metrizable, and is frequently a group-invariant topological model for suitable random walks on the associated group. In comparison, Morse boundary, constructed in [ Cor18 , CS15 ] , in comparison, often lacks metrizability and Poisson boundary realization property. Meanwhile, the genericity of a more geometric flavor is also exhibited for sublinearly Morse boundaries. In [ GQR22 ] , genericity of sublinearly Morse directions under Patterson Sullivan measure was shown to hold in the more general context of actions which admit a strongly contracting element. In fact, the results in [ GQR22 ] concerning stationary measures were recently claimed in a different setting by Inhyeok Choi [ Choi22 ] , who, in place of ergodic theoretic and boundary techniques uses a pivoting technique developed by Gouëzel [ Gou22 ] . Also, following [ Yan22 ] , genericity of sublinearly Morse directions on the horofunction boundary was shown for all proper statistically convex-cocompact actions on proper metric spaces [ QY24 ] . Most recently, Wen defined and studied the genericity sublinearly Morse directions in higher rank symmetric spaces [ Wen25 ] . Open questions An immediate question to ask is if the sublinearly Morse property can be detected on the bi-infinite geodesic line exhibited in this paper. The authors of this paper thus continue to study and aim to give a positive answer in an upcoming preprint [ JQ ] . A harder and open question is to ask if first passage percolations studied in this paper preserve the Morse property of the bi-infinite geodesic line. 2. preliminaries 2.1. Quasi-isometry and quasi-isometric embeddings Definition 2.1 (Quasi Isometric embedding) . Let (X,d_{X}) and (Y,d_{Y}) be metric spaces. For constants q\geq 1 and Q\geq 0 , we say a map f\colon\thinspace X\to Y is a (q,Q) – quasi-isometric embedding if, for all points x_{1},x_{2}\in X \frac{1}{q}d_{X}(x_{1},x_{2})-Q\leq

Read the complete text