Optimal One-Pass Nonparametric Estimation Under Memory Constraint

DOI10.1080/01621459.2022.2115374MaRDI QIDQ6153992zbMATH OpenOpenAlexFDO

Authors Mingxue Quan, Zhenhua Lin

Publication date 19 March 2024

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

Full work available at URL https://arxiv.org/abs/2208.08668, https://figshare.com/articles/dataset/Optimal_One-pass_Nonparametric_Estimation_Under_Memory_Constraint/20536989

zbMATH Keywords

nonparametric regressionstreaming algorithmcommunication complexityminimaxityspace lower bound

Mathematics Subject Classification ID

Statistics (62-XX)

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.

Cites work

• scientific article; zbMATH DE number 3949496 (Why is no real title available?)
• scientific article; zbMATH DE number 3986385 (Why is no real title available?)
• scientific article; zbMATH DE number 45848 (Why is no real title available?)
• scientific article; zbMATH DE number 47282 (Why is no real title available?)
• scientific article; zbMATH DE number 700016 (Why is no real title available?)
• scientific article; zbMATH DE number 3222478 (Why is no real title available?)
• Communication Complexity
• Dynamic Penalized Splines for Streaming Data
• How to apply the method of stochastic approximation in the non-parametric estimation of a regression function¹
• Local linear regression smoothers and their minimax efficiencies
• Minimax estimation via wavelet shrinkage
• On randomized one-round communication complexity
• On the communication complexity of distributed algebraic computation
• Optimal global rates of convergence for nonparametric regression
• Optimal uniform convergence rates and asymptotic normality for series estimators under weak dependence and weak conditions
• Recursive Nonparametric Estimation for Time Series
• Recursive local polynomial regression under dependence conditions
• Revisiting R\'ev\'esz's stochastic approximation method for the estimation of a regression function
• Some new asymptotic theory for least squares series: pointwise and uniform results
• Spectral Methods
• Wavelets on the interval and fast wavelet transforms

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)