Distributed statistical estimation and rates of convergence in normal approximation

DOI10.1214/19-EJS1647MaRDI QIDQ2283576zbMATHFDO

Authors Stanislav Minsker

Publication date 3 January 2020

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1704.02658

zbMATH Keywords

normal approximationrobust estimationlimit theoremsmedian-of-means estimatordistributed estimation

Mathematics Subject Classification ID

Density estimation (62G07) Nonparametric robustness (62G35) Functional limit theorems; invariance principles (60F17)

Abstract: This paper presents a class of new algorithms for distributed statistical estimation that exploit divide-and-conquer approach. We show that one of the key benefits of the divide-and-conquer strategy is robustness, an important characteristic for large distributed systems. We establish connections between performance of these distributed algorithms and the rates of convergence in normal approximation, and prove non-asymptotic deviations guarantees, as well as limit theorems, for the resulting estimators. Our techniques are illustrated through several examples: in particular, we obtain new results for the median-of-means estimator, as well as provide performance guarantees for distributed maximum likelihood estimation.

Recommendations

• Distributed testing and estimation under sparse high dimensional models
• Distributed statistical inference for massive data
• Distributed nonparametric function estimation: optimal rate of convergence and cost of adaptation
• Communication-efficient algorithms for statistical optimization
• An asymptotic analysis of distributed nonparametric methods

Cites work

• scientific article; zbMATH DE number 3954047 (Why is no real title available?)
• scientific article; zbMATH DE number 3790208 (Why is no real title available?)
• scientific article; zbMATH DE number 49190 (Why is no real title available?)
• scientific article; zbMATH DE number 1256715 (Why is no real title available?)
• scientific article; zbMATH DE number 741240 (Why is no real title available?)
• scientific article; zbMATH DE number 5058453 (Why is no real title available?)
• A Berry–Esséen Bound for M‐estimators
• A Class of Statistics with Asymptotically Normal Distribution
• Approximation Theorems of Mathematical Statistics
• Bandits With Heavy Tail
• Challenging the empirical mean and empirical variance: a deviation study
• Distributed estimation of principal eigenspaces
• Estimates of Location Based on Rank Tests
• Geometric median and robust estimation in Banach spaces
• Learning from untrusted data
• Loss minimization and parameter estimation with heavy tails
• Near-optimal mean estimators with respect to general norms
• Note on the median of a multivariate distribution
• On Some Robust Estimates of Location
• On statistics, computation and scalability
• On the Medians of Gamma Distributions and an Equation of Ramanujan
• On the estimation of the mean of a random vector
• On the optimality of averaging in distributed statistical learning
• Probability Inequalities for Sums of Bounded Random Variables
• Random generation of combinatorial structures from a uniform distribution
• Robust Estimation of a Location Parameter
• Robust and scalable Bayes via a median of subset posterior measures
• Simple, scalable and accurate posterior interval estimation
• Sub-Gaussian estimators of the mean of a random vector
• Sub-Gaussian mean estimators
• The Bahadur-Kiefer representation for \(U\)-quantiles

Cited in

(27)

• A review of distributed statistical inference
• Byzantine-robust distributed sparse learning for \(M\)-estimation
• Discussion of the paper ‘A review of distributed statistical inference’
• Discussion of: ‘A review of distributed statistical inference’
• A massive data framework for M-estimators with cubic-rate
• Robust sub-Gaussian estimation of a mean vector in nearly linear time
• Distributed testing and estimation under sparse high dimensional models
• Distributed nonparametric function estimation: optimal rate of convergence and cost of adaptation
• The statistical performance of collaborative inference
• Distributed subsampling for multiplicative regression
• The COR criterion for optimal subset selection in distributed estimation
• Nonparametric distributed learning under general designs
• Large-Scale Inference of Multivariate Regression for Heavy-Tailed and Asymmetric Data
• Robust subgaussian estimation with VC-dimension
• Emerging directions in Bayesian computation
• Design of distributed BFGS algorithm and large sample properties for parameter estimation
• Robust classification via MOM minimization
• Individual Data Protected Integrative Regression Analysis of High-Dimensional Heterogeneous Data
• LIC criterion for optimal subset selection in distributed interval estimation
• scientific article; zbMATH DE number 7470640 (Why is no real title available?)
• Robust and efficient mean estimation: an approach based on the properties of self-normalized sums
• scientific article; zbMATH DE number 7370601 (Why is no real title available?)
• Robust machine learning by median-of-means: theory and practice
• Distributed Bayesian inference in massive spatial data
• An asymptotic analysis of distributed nonparametric methods
• Robust distributed multicategory angle-based classification for massive data
• First-Order Newton-Type Estimator for Distributed Estimation and Inference

This page was built for publication: Distributed statistical estimation and rates of convergence in normal approximation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2283576)