Optimal One-Pass Nonparametric Estimation Under Memory Constraint

DOI10.1080/01621459.2022.2115374arXiv2208.08668OpenAlexW4292657694MaRDI QIDQ6153992FDOQ6153992

Authors: Mingxue Quan, Zhenhua Lin 

Publication date: 19 March 2024

Published in: Journal of the American Statistical Association (Search for Journal in Brave)

Abstract: For nonparametric regression in the streaming setting, where data constantly flow in and require real-time analysis, a main challenge is that data are cleared from the computer system once processed due to limited computer memory and storage. We tackle the challenge by proposing a novel one-pass estimator based on penalized orthogonal basis expansions and developing a general framework to study the interplay between statistical efficiency and memory consumption of estimators. We show that, the proposed estimator is statistically optimal under memory constraint, and has asymptotically minimal memory footprints among all one-pass estimators of the same estimation quality. Numerical studies demonstrate that the proposed one-pass estimator is nearly as efficient as its non-streaming counterpart that has access to all historical data.

Full work available at URL: https://arxiv.org/abs/2208.08668

zbMATH Keywords

nonparametric regressionstreaming algorithmcommunication complexityminimaxityspace lower bound

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Local linear regression smoothers and their minimax efficiencies
• Title not available (Why is that?)
• Title not available (Why is that?)
• Optimal global rates of convergence for nonparametric regression
• Wavelets on the interval and fast wavelet transforms
• Optimal uniform convergence rates and asymptotic normality for series estimators under weak dependence and weak conditions
• Some new asymptotic theory for least squares series: pointwise and uniform results
• Spectral Methods
• Minimax estimation via wavelet shrinkage
• On randomized one-round communication complexity
• Title not available (Why is that?)
• Recursive local polynomial regression under dependence conditions
• How to apply the method of stochastic approximation in the non-parametric estimation of a regression function¹
• Recursive Nonparametric Estimation for Time Series
• Revisiting R\'ev\'esz's stochastic approximation method for the estimation of a regression function
• Communication Complexity
• On the communication complexity of distributed algebraic computation
• Dynamic Penalized Splines for Streaming Data

Cited In (2)

• Renewable Huber estimation method for streaming datasets
• Non-crossing quantile double-autoregression for the analysis of streaming time series data

This page was built for publication: Optimal One-Pass Nonparametric Estimation Under Memory Constraint

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6153992)