Id: arxiv/2608.01182 Url: https://arxiv.org/abs/2608.01182 Tags: math.AP, math.PR, stat.ML Score: 0 Title: Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels Domain: math.AP Source: arxiv Authors: Matthew Rosenzweig, Dejan Slepčev, Lihan Wang Reviews: 0 Subject: math.AP Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-08-02 Page: https://delphi.science/r/arxiv/2608.01182 We study the Wasserstein gradient flow of the squared Maximum Mean Discrepancy (MMD) generated by the nonsmooth energy kernels $K(z)=-|z|^q$, $0 0$, we prove global well-posedness on $\mathbb{R}^d$ for probability densities in subcritical $L^p$ spaces, with targets in the same integrability class and with finite moments. We also include the one-dimensional Coulomb endpoint $d=q=1$. For the associated $N$-particle system, we prove global noncollision and fixed-$N$ convergence to the collision-free critical set, a particle-to-continuum criticality principle, and a modulated-energy mean-field…