Efficient median of means estimator

2023/05/30 by Stanislav Minsker, Minsker, Stanislav · 5 citations
Mathematics · Computer Science · #Statistical Methods and Inference #Advanced Statistical Methods and Models #Bayesian Methods and Mixture Models

paper · pdf · doi:10.48550/arxiv.2305.18681

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 by the author, and prove that desired guarantees can be attained under weaker requirements.

Cited by

• Universal Refinement without Interaction: Order-Optimal 1-Bit Mean Estimation
• Improved Concentration for Mean Estimators via Shrinkage
• On the Optimality of the Median-of-Means Estimator under Adversarial Contamination
• Sequential 1-bit Mean Estimation with Near-Optimal Sample Complexity
• Adversarially robust multiple testing in high dimensions

Related

• U-statistics of growing order and sub-Gaussian mean estimators with sharp constants
• Optimality in Mean Estimation: Beyond Worst-Case, Beyond Sub-Gaussian, and Beyond 1+α Moments
• Optimal Sub-Gaussian Mean Estimation in ℝ
• Robust machine learning by median-of-means: Theory and practice
• Sub-Gaussian estimators of the mean of a random vector