Space-time boundaries for random walks and their application to operator algebras
Adam Dor-On, Ilya Gekhtman, Pavel Prudnikov

TL;DR
This paper explores the space-time Martin boundary of random walks on groups, relating it to classical compactifications, and applies these findings to operator algebras, revealing structural insights and boundaries.
Contribution
It introduces the $0$-Martin boundary, relates it to classical compactifications, and identifies the minimal space-time Martin boundary with a union of $ ext{Martin}$ boundaries, connecting probabilistic and operator algebraic structures.
Findings
The reduced ratio-limit compactification embeds into the space-time Martin boundary.
The $0$-Martin boundary governs $ ext{infty}$-harmonic functions and arises as limits of $ ext{Martin}$ kernels.
The noncommutative Shilov boundary coincides with the Toeplitz $C^*$-algebra.
Abstract
We investigate the Martin boundary of the space-time Markov chain associated to a finitely supported random walk with spectral radius and relate it to several classical compactifications of . Assuming the strong ratio-limit property, we prove that the reduced ratio-limit compactification embeds naturally into the space-time Martin boundary. We introduce the -Martin boundary, which governs the behaviour of -harmonic functions, and show that the -Martin kernels arise as rescaled limits of -Martin kernels as . For symmetric random walks on hyperbolic groups, the -Martin boundary naturally covers the Gromov boundary, while the cover need not be injective in general. Our main structural theorem identifies the minimal space-time Martin boundary with the disjoint union of minimal -Martin boundaries over…
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
TopicsAdvanced Operator Algebra Research · Geometric and Algebraic Topology · Markov Chains and Monte Carlo Methods
