Near-Lipschitz stability of the Kim--Milman flow map
Sinho Chewi, Katharina Eichinger, Aram-Alexandre Pooladian
math.PR, math.FA, stat.ML
Submitted: 2026-06-22
Comments: 28 pages
License: http://creativecommons.org/licenses/by/4.0/
The gist: We prove that the Kim--Milman flow map enjoys favorable stability properties with respect to variations in the target measure, provided that one of the target measures is sufficiently regular.
Terminology
Abstract
We prove that the Kim--Milman flow map enjoys favorable stability properties with respect to variations in the target measure, provided that one of the target measures is sufficiently regular. Our results include stability in relative entropy, and more notably, Lipschitz stability in the 2-Wasserstein distance up to a logarithmic factor. We complement our results with a general existence theorem for these maps for any target measure with finite second moment.
Sources
- Stability Bounds for Smooth Optimal Transport Maps and their Statistical Implications
- Statistical Estimation of Monge Transport Maps via Brenier Potentials
- Propagation of weak log-concavity along generalised heat flows via Hamilton-Jacobi equations
- Stability of the Kim--Milman flow map
- A coupling approach to Lipschitz transport maps
- A Generalization of Caffarelli's Contraction Theorem to Nearly Spherical Manifolds
- Global Regularity Estimates for Optimal Transport via Entropic Regularisation
- Stability of optimal transport maps on Riemannian manifolds
- Gluing methods for quantitative stability of optimal transport maps
- Lipschitz changes of variables via heat flow
- Entropic estimation of optimal transport maps
Related papers
- Sharp Deviations Bounds for Dirichlet Weighted Sums with Application to analysis of Bayesian algorithms
- Subspace Uncertainty and Sharp Sampling Thresholds on the Boolean Cube
- Local Anticoncentration for Gaussian Boson Sampling via Conditional Wishart Geometry
- Beyond the Semicircle: Free Diffusion Models with Prescribed Equilibria
- Explicit Steady-State Approximations for Parallel Server Systems with Heterogeneous Servers
- A New Bound on the Cumulant Generating Function of Dirichlet Processes