MLLGMay 30

On Median of Incomplete U-Statistics

arXiv:2606.0066112.9h-index: 3
Predicted impact top 46% in ML · last 90 daysOriginality Incremental advance
AI Analysis

Provides a theoretical guarantee for a robust estimator of symmetric kernel expectations, relevant to statistical learning and nonparametric estimation.

The paper establishes the finite-sample concentration rate for the Median-of-Incomplete-U-Statistics (MIU), showing it achieves exponential concentration with rate n^{-1/2} for kernels of any order, improving over existing results.

We establish the finite-sample concentration rate for the Median-of-Incomplete-U-Statistics (MIU), an efficient robust estimator for the expectation of symmetric kernels.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes