paper

Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels

arXiv:2608.01182

Abstract

We study the Wasserstein gradient flow of the squared Maximum Mean Discrepancy (MMD) generated by the nonsmooth energy kernels , . In dimensions , the corresponding energies are not displacement semiconvex, so standard Wasserstein-gradient-flow theory does not apply. When , we prove global well-posedness on for probability densities in subcritical spaces, with targets in the same integrability class and with finite moments. We also include the one-dimensional Coulomb endpoint . For the associated -particle system, we prove global noncollision and fixed- convergence to the collision-free critical set, a particle-to-continuum criticality principle, and a modulated-energy mean-field estimate that yields convergence of the particle dynamics to the continuum flow as on every finite time interval. We also construct collision-free saddle equilibria, showing that deterministic particle trajectories need not approach global empirical minimizers. For , every continuum solution in our class has a narrowly relatively compact orbit, every -limit point is Lagrangian critical, and the orbit approaches the Lagrangian critical set. For , the same conclusions hold under uniform-in-time moment and subcritical bounds. We prove that an absolutely continuous Lagrangian critical point equals the target when the source and target have finite moments of order , except when and . Under the preceding uniform bounds, rigidity gives convergence of the continuum flow to the target throughout the rigid part of the well-posedness range. Finally, we show that no initial-data-independent multiplicative MMD decay modulus exists on , and that global Polyak--Łojasiewicz inequalities fail in several whole-space and periodic Riesz/Coulomb regimes.

Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels · wovepaper