Stationary MMD Points
Zonghao Chen, Toni Karvonen, Heishiro Kanagawa, François-Xavier Briol, Chris J. Oates · 2025
Read Stationary MMD Points by Zonghao Chen, Toni Karvonen, Heishiro Kanagawa, François-Xavier Briol, Chris J. Oates (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-20754.txt.
Description
Zonghao Chen and co-authors study stationary points of the maximum mean discrepancy, a tool for summing up a probability law with a finite set of points. Such point sets serve numeric integration, and many authors pick them by shrinking the discrepancy. The catch is that the goal is non-convex, so a global best point set is hard to find.
Their 2025 paper turns to stationary points, which can be computed with care. For test functions in the linked Hilbert space, the integration error at such points shrinks faster than the discrepancy itself, a super-convergence effect. The authors then show that gradient flow on the discrepancy can reach these points, with a new finite-particle error bound.
Catalog record
| CREATOR | Zonghao Chen, Toni Karvonen, Heishiro Kanagawa, François-Xavier Briol, Chris J. Oates |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 114 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-20754