Efficient Median of Means Estimator

Abstract

The goal of this note is to present a modification of the popular median of means estimator that achieves sub-Gaussian deviation bounds with nearly optimal constants under minimal assumptions on the underlying distribution. We build on a recent work on the topic and prove that desired guarantees can be attained under weaker requirements.

Cite

Text

Minsker. "Efficient Median of Means Estimator." Conference on Learning Theory, 2023.

Markdown

[Minsker. "Efficient Median of Means Estimator." Conference on Learning Theory, 2023.](https://mlanthology.org/colt/2023/minsker2023colt-efficient/)

BibTeX

@inproceedings{minsker2023colt-efficient,
  title     = {{Efficient Median of Means Estimator}},
  author    = {Minsker, Stanislav},
  booktitle = {Conference on Learning Theory},
  year      = {2023},
  pages     = {5925-5933},
  volume    = {195},
  url       = {https://mlanthology.org/colt/2023/minsker2023colt-efficient/}
}