Adaptive distributed methods under communication constraints

DOI10.1214/19-AOS1890MaRDI QIDQ2215741zbMATH OpenFDO

Authors Harry van Zanten, Botond Szabó

Publication date 14 December 2020

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1804.00864, https://projecteuclid.org/euclid.aos/1597370676

zbMATH Keywords

adaptationcommunication constraintsdistributed computationminimax ratesnonparametric regression

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Minimax procedures in statistical decision theory (62C20) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20)

Abstract: We study distributed estimation methods under communication constraints in a distributed version of the nonparametric random design regression model. We derive minimax lower bounds and exhibit methods that attain those bounds. Moreover, we show that adaptive estimation is possible in this setting.

Recommendations

• Distributed nonparametric function estimation: optimal rate of convergence and cost of adaptation
• Communication lower bounds for statistical estimation problems via a distributed data processing inequality
• An asymptotic analysis of distributed nonparametric methods
• Distributed sparse normal means estimation with sublinear communication
• Distributed estimation and quantization

Cites work

• A Scalable Bootstrap for Massive Data
• Adaptive estimation of the transition density of a particular hidden Markov chain
• An asymptotic analysis of distributed nonparametric methods
• Communication-efficient sparse regression
• Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
• Distributed testing and estimation under sparse high dimensional models
• Honest and adaptive confidence sets in L_p
• Information-theoretic determination of minimax rates of convergence
• Introduction to nonparametric estimation
• Mathematical foundations of infinite-dimensional statistical models
• On the optimality of averaging in distributed statistical learning
• On the Sum Rate of Gaussian Multiterminal Source Coding: New Proofs and Results
• PenalizationversusGoldenshluger − Lepski strategies in warped bases regression
• Quantized minimax estimation over Sobolev ellipsoids
• Wavelets, approximation, and statistical applications

Cited in

(16)

• Distributed linear regression by averaging
• Distributed adaptive Gaussian mean estimation with unknown variance: interactive protocol helps adaptation
• Distributed nonparametric function estimation: optimal rate of convergence and cost of adaptation
• Distributed function estimation: adaptation using minimal communication
• Consistency in models for distributed learning under communication constraints
• Distributed Decision-Making Over Adaptive Networks
• Learning Theory
• Distributed sparse normal means estimation with sublinear communication
• An asymptotic analysis of distributed nonparametric methods
• Nonparametric Bayesian aggregation for massive data
• Quantized minimax estimation over Sobolev ellipsoids
• Communication lower bounds for statistical estimation problems via a distributed data processing inequality
• Distributed adaptive consensus protocol with decaying gains
• Optimal high-dimensional and nonparametric distributed testing under communication constraints
• On lower bounds for the bias-variance trade-off
• Efficient Quantization Mean Estimation for Distributed Learning

This page was built for publication: Adaptive distributed methods under communication constraints

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2215741)