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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Diakonikolas, Ilias, Kane, Daniel M., Karmalkar, Sushrut, Pensia, Ankit, Pittas, Thanasis
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910513120673792
author Diakonikolas, Ilias
Kane, Daniel M.
Karmalkar, Sushrut
Pensia, Ankit
Pittas, Thanasis
author_facet Diakonikolas, Ilias
Kane, Daniel M.
Karmalkar, Sushrut
Pensia, Ankit
Pittas, Thanasis
contents We study the problem of list-decodable sparse mean estimation. Specifically, for a parameter $α\in (0, 1/2)$, we are given $m$ points in $\mathbb{R}^n$, $\lfloor αm \rfloor$ 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 $\widehat μ$ such that $\| \widehat μ- μ\|_2$ 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 $m = (k\log(n))^{O(t)}/α$ and running time $\mathrm{poly}(mn^t)$. For the special case of Gaussian inliers, our algorithm achieves the optimal error guarantee of $Θ(\sqrt{\log(1/α)})$ with quasi-polynomial sample and computational complexity. We complement our upper bounds with nearly-matching statistical query and low-degree polynomial testing lower bounds.
format Preprint
id arxiv_https___arxiv_org_abs_2206_05245
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle List-Decodable Sparse Mean Estimation via Difference-of-Pairs Filtering
Diakonikolas, Ilias
Kane, Daniel M.
Karmalkar, Sushrut
Pensia, Ankit
Pittas, Thanasis
Data Structures and Algorithms
Machine Learning
Statistics Theory
We study the problem of list-decodable sparse mean estimation. Specifically, for a parameter $α\in (0, 1/2)$, we are given $m$ points in $\mathbb{R}^n$, $\lfloor αm \rfloor$ 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 $\widehat μ$ such that $\| \widehat μ- μ\|_2$ 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 $m = (k\log(n))^{O(t)}/α$ and running time $\mathrm{poly}(mn^t)$. For the special case of Gaussian inliers, our algorithm achieves the optimal error guarantee of $Θ(\sqrt{\log(1/α)})$ with quasi-polynomial sample and computational complexity. We complement our upper bounds with nearly-matching statistical query and low-degree polynomial testing lower bounds.
title List-Decodable Sparse Mean Estimation via Difference-of-Pairs Filtering
topic Data Structures and Algorithms
Machine Learning
Statistics Theory
url https://arxiv.org/abs/2206.05245