Langevin deformation for R\'enyi entropy on Wasserstein space over Riemannian manifolds
Rong Lei, Songzi Li, Xiang-Dong Li

TL;DR
This paper introduces a Langevin deformation for R'enyi entropy on Wasserstein space over Riemannian manifolds, unifying several flow equations and establishing entropy, monotonicity, and convergence properties with new results even in Euclidean and non-negatively curved spaces.
Contribution
It develops a novel Langevin deformation framework for R'enyi entropy on Wasserstein space, connecting porous medium and geodesic flows, and proves new entropy, monotonicity, and convergence results.
Findings
Proved $W$-entropy-information formulae and rigidity theorems.
Established monotonicity of the Hamiltonian and convexity of the Lagrangian.
Demonstrated convergence of the deformation as parameters tend to zero or infinity.
Abstract
We introduce the Langevin deformation for the R\'enyi entropy on the -Wasserstein space over or a Riemannian manifold, which interpolates between the porous medium equation and the Benamou-Brenier geodesic flow on the -Wasserstein space and can be regarded as the compressible Euler equations for isentropic gas with damping. We prove the -entropy-information formulae and the the rigidity theorems for the Langevin deformation for the R\'enyi entropy on the Wasserstein space over complete Riemannian manifolds with non-negative Ricci curvature or CD-condition. Moreover, we prove the monotonicity of the Hamiltonian and the convexity of the Lagrangian along the Langevin deformation of flows. Finally, we prove the convergence of the Langevin deformation for the R\'enyi entropy as and respectively. Our results are new…
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
