A non-asymptotic distributional theory of approximate message passing for sparse and robust regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Gen, Wei, Yuting
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914634180591616
author Li, Gen
Wei, Yuting
author_facet Li, Gen
Wei, Yuting
contents Characterizing the distribution of high-dimensional statistical estimators is a challenging task, due to the breakdown of classical asymptotic theory in high dimension. This paper makes progress towards this by developing non-asymptotic distributional characterizations for approximate message passing (AMP) -- a family of iterative algorithms that prove effective as both fast estimators and powerful theoretical machinery -- for both sparse and robust regression. Prior AMP theory, which focused on high-dimensional asymptotics for the most part, failed to describe the behavior of AMP when the number of iterations exceeds $o\big({\log n}/{\log \log n}\big)$ (with $n$ the sample size). We establish the first finite-sample non-asymptotic distributional theory of AMP for both sparse and robust regression that accommodates a polynomial number of iterations. Our results derive approximate accuracy of Gaussian approximation of the AMP iterates, which improves upon all prior results and implies enhanced distributional characterizations for both optimally tuned Lasso and robust M-estimator.
format Preprint
id arxiv_https___arxiv_org_abs_2401_03923
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A non-asymptotic distributional theory of approximate message passing for sparse and robust regression
Li, Gen
Wei, Yuting
Statistics Theory
Information Theory
Machine Learning
Signal Processing
Characterizing the distribution of high-dimensional statistical estimators is a challenging task, due to the breakdown of classical asymptotic theory in high dimension. This paper makes progress towards this by developing non-asymptotic distributional characterizations for approximate message passing (AMP) -- a family of iterative algorithms that prove effective as both fast estimators and powerful theoretical machinery -- for both sparse and robust regression. Prior AMP theory, which focused on high-dimensional asymptotics for the most part, failed to describe the behavior of AMP when the number of iterations exceeds $o\big({\log n}/{\log \log n}\big)$ (with $n$ the sample size). We establish the first finite-sample non-asymptotic distributional theory of AMP for both sparse and robust regression that accommodates a polynomial number of iterations. Our results derive approximate accuracy of Gaussian approximation of the AMP iterates, which improves upon all prior results and implies enhanced distributional characterizations for both optimally tuned Lasso and robust M-estimator.
title A non-asymptotic distributional theory of approximate message passing for sparse and robust regression
topic Statistics Theory
Information Theory
Machine Learning
Signal Processing
url https://arxiv.org/abs/2401.03923