Sub-Gaussian Mean Estimation in Polynomial Time

09/19/2018
by   Samuel B. Hopkins, et al.
0

We study polynomial time algorithms for estimating the mean of a random vector X in R^d from n independent samples X_1,...,X_n when X may be heavy-tailed. We assume only that X has finite mean μ and covariance Σ. In this setting, the radius of confidence intervals achieved by the empirical mean are large compared to the case that X is Gaussian or sub-Gaussian. In particular, for confidence δ > 0, the empirical mean has confidence intervals with radius of order √(TrΣ / δ n) rather than √(TrΣ /n ) + √(λ_(Σ) (1/δ) / n) from the Gaussian case. We offer the first polynomial time algorithm to estimate the mean with sub-Gaussian confidence intervals under such mild assumptions. Our algorithm is based on a new semidefinite programming relaxation of a high-dimensional median. Previous estimators which assumed only existence of O(1) moments of X either sacrifice sub-Gaussian performance or are only known to be computable via brute-force search procedures requiring (d) time.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/19/2018

Mean Estimation with Sub-Gaussian Rates in Polynomial Time

We study polynomial time algorithms for estimating the mean of a heavy-t...
research
12/23/2019

Algorithms for Heavy-Tailed Statistics: Regression, Covariance Estimation, and Beyond

We study efficient algorithms for linear regression and covariance estim...
research
02/06/2019

Fast Mean Estimation with Sub-Gaussian Rates

We propose an estimator for the mean of a random vector in R^d that can ...
research
12/30/2017

New robust confidence intervals for the mean under dependence

The goal of this paper is to indicate a new method for constructing norm...
research
12/20/2022

A polynomial time additive estimate of the permanent using Gaussian fields

We present a polynomial-time randomized algorithm for estimating the per...
research
12/10/2018

Testing for high-dimensional network paramaters in auto-regressive models

High-dimensional auto-regressive models provide a natural way to model i...
research
06/18/2020

A Framework for Sample Efficient Interval Estimation with Control Variates

We consider the problem of estimating confidence intervals for the mean ...

Please sign up or login with your details

Forgot password? Click here to reset