Recurrence of horizontal-vertical walks
Swee Hong Chan

TL;DR
This paper studies a randomized 2D lattice walk with label-dependent directions, showing recurrence for certain resampling probabilities, advancing understanding of related deterministic rotor walks whose recurrence remains unresolved.
Contribution
It proves recurrence of the horizontal-vertical walk for resampling probability q in (1/3, 1], providing insights into the behavior of similar deterministic models.
Findings
The walk is recurrent for q in (1/3, 1].
Recurrence holds with probability 1 for these parameters.
The result advances understanding of rotor walk behavior.
Abstract
Consider a nearest neighbor random walk on the two-dimensional integer lattice, where each vertex is initially labeled either `H' or `V', uniformly and independently. At each discrete time step, the walker resamples the label at its current location (changing `H' to `V' and `V' to `H' with probability ). Then, it takes a mean zero horizontal step if the new label is `H', and a mean zero vertical step if the new label is `V'. This model is a randomized version of the deterministic rotor walk, for which its recurrence (i.e., visiting every vertex infinitely often with probability 1) in two dimensions is still an open problem. We answer the analogous question for the the horizontal-vertical walk, by showing that the horizontal-vertical walk is recurrent for .
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsStochastic processes and statistical mechanics · Advanced Combinatorial Mathematics · Markov Chains and Monte Carlo Methods
