The Minimax Lower Bound of Kernel Stein Discrepancy Estimation

May 2026 · Papers

AISTATS 2026

We studied the minimax lower bound of Kernel Stein Discrepancy (KSD) estimation in $\mathbb R^d$ and in general topological spaces. We found that the fastest estimation for the KSD one can achieve is $\mathcal O(n^{-1/2})$.

Cite
@INPROCEEDINGS{cribeiro26minimax,
  AUTHOR =       {Jose Cribeiro-Ramallo and Agnideep Aich and Florian Kalinke and Ashit Baran Aich and Zolt{\'a}n Szab{\'o}},
  TITLE =        {The Minimax Lower Bound of Kernel {S}tein Discrepancy Estimation},
  BOOKTITLE =    {International Conference on Artificial Intelligence and Statistics (AISTATS)},
  YEAR =         {2026},
  pages =        {901--909},
}