List-Decodable Sparse Mean Estimation via Difference-of-Pairs Filtering

06/10/2022
by   Ilias Diakonikolas, et al.
5

We study the problem of list-decodable sparse mean estimation. Specifically, for a parameter α∈ (0, 1/2), we are given m points in ℝ^n, ⌊α m ⌋ of which are i.i.d. samples from a distribution D with unknown k-sparse mean μ. No assumptions are made on the remaining points, which form the majority of the dataset. The goal is to return a small list of candidates containing a vector μ such that μ- μ_2 is small. Prior work had studied the problem of list-decodable mean estimation in the dense setting. In this work, we develop a novel, conceptually simpler technique for list-decodable mean estimation. As the main application of our approach, we provide the first sample and computationally efficient algorithm for list-decodable sparse mean estimation. In particular, for distributions with “certifiably bounded” t-th moments in k-sparse directions and sufficiently light tails, our algorithm achieves error of (1/α)^O(1/t) with sample complexity m = (klog(n))^O(t)/α and running time poly(mn^t). For the special case of Gaussian inliers, our algorithm achieves the optimal error guarantee of Θ (√(log(1/α))) with quasi-polynomial sample and computational complexity. We complement our upper bounds with nearly-matching statistical query and low-degree polynomial testing lower bounds.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/07/2022

Robust Sparse Mean Estimation via Sum of Squares

We study the problem of high-dimensional sparse mean estimation in the p...
research
05/28/2022

List-Decodable Sparse Mean Estimation

Robust mean estimation is one of the most important problems in statisti...
research
11/07/2016

Learning from Untrusted Data

The vast majority of theoretical results in machine learning and statist...
research
06/18/2020

List-Decodable Mean Estimation via Iterative Multi-Fitering

We study the problem of list-decodable mean estimation for bounded cova...
research
07/10/2020

Learning Entangled Single-Sample Gaussians in the Subset-of-Signals Model

In the setting of entangled single-sample distributions, the goal is to ...
research
05/04/2020

Learning Strong Substitutes Demand via Queries

This paper addresses the computational challenges of learning strong sub...
research
06/18/2020

List-Decodable Mean Estimation via Iterative Multi-Filtering

We study the problem of list-decodable mean estimation for bounded covar...

Please sign up or login with your details

Forgot password? Click here to reset