Non-Markovian maximal couplings and a vertical reflection principle on a class of sub-Riemannian manifolds
Liangbing Luo, Robert W. Neel

TL;DR
This paper introduces a novel class of non-Markovian couplings for sub-Riemannian Brownian motions on certain manifolds, leveraging global isometries to achieve maximal couplings with a reflection principle, improving bounds on coupling times.
Contribution
It develops a unified, simple method for constructing maximal, non-Markovian couplings on various sub-Riemannian manifolds using global isometries, extending prior work and providing explicit coupling time bounds.
Findings
Couplings are maximal and based on global isometries.
Coupling time reduces to hitting time of a Brownian component.
Applications include bounds for heat semigroup inequalities.
Abstract
We develop an approach to constructing non-Markovian, non-co-adapted couplings for sub-Riemannian Brownian motions in sub-Riemannian manifolds with large symmetry groups by treating the specific cases of the three-dimensional Heisenberg group, higher-dimensional non-isotropic Heisenberg groups, SL(2,R) and its universal cover, and SU(2). Our primary focus is on the situation when the processes start from two points on the same vertical fiber, since in general Markovian or co-adapted couplings cannot give the sharp rate for the coupling time in this case. Non-Markovian couplings of this type on sub-Riemannian manifolds were first introduced by Banerjee-Gordina-Mariano, for the three-dimensional Heisenberg group, and were more recently extended by B\'en\'efice to SL(2,R) and SU(2), using a detailed consideration of the Brownian bridge. In contrast, our couplings are based on global…
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
TopicsGeometric Analysis and Curvature Flows · Topological and Geometric Data Analysis · Mathematical Dynamics and Fractals
