no code implementations • 30 May 2023 • Stanislav Minsker
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.
no code implementations • 13 Nov 2021 • Stanislav Minsker, Mohamed Ndaoud, Yiqiu Shen
Our analysis also shows that interpolation can be robust to corruption in the covariance of the noise when the signal is aligned with the "clean" part of the covariance, for the properly defined notion of alignment.
no code implementations • 5 Apr 2020 • Stanislav Minsker
This paper investigates asymptotic properties of algorithms that can be viewed as robust analogues of the classical empirical risk minimization.
no code implementations • 16 Oct 2019 • Stanislav Minsker, Timothée Mathieu
We propose a version of empirical risk minimization based on the idea of replacing sample averages by robust proxies of the expectation, and obtain high-confidence bounds for the excess risk of resulting estimators.
1 code implementation • 5 Nov 2018 • Yuan Ke, Stanislav Minsker, Zhao Ren, Qiang Sun, Wen-Xin Zhou
We offer a survey of recent results on covariance estimation for heavy-tailed distributions.
Methodology Statistics Theory Statistics Theory
no code implementations • 9 Apr 2017 • Stanislav Minsker, Nate Strawn
This paper presents a class of new algorithms for distributed statistical estimation that exploit divide-and-conquer approach.
no code implementations • 23 May 2016 • Stanislav Minsker
As is now well known, the sample mean then may have a catastrophically bad performance..." Motivated by this question, we develop a new estimator of the (element-wise) mean of a random matrix, which includes covariance estimation problem as a special case.
no code implementations • 11 Mar 2014 • Stanislav Minsker, Sanvesh Srivastava, Lizhen Lin, David B. Dunson
We propose a novel approach to Bayesian analysis that is provably robust to outliers in the data and often has computational advantages over standard methods.