Winsorized mean estimation with heavy tails and adversarial contamination
Abstract
Finite-sample upper bounds on the estimation error of a winsorized mean estimator of the population mean in the presence of heavy tails and adversarial contamination are established. In comparison to existing results, the winsorized mean estimator we study avoids a sample-splitting device and winsorizes substantially fewer observations, which improves its applicability and practical performance.
Fields of Study & Subjects
Adversarial Robustness in Machine Learning
Advanced Statistical Methods and Models
Anomaly Detection Techniques and Applications
Publication details
- Journal / source
- Electronic Journal of Statistics
- Volume
- 20
- Issue
- 2
- Pages
- 3225–3254
- Publication date
- 1 January 2026
- Language
- EN