Distributed Mean Estimation with Optimal Error Bounds

02/21/2020
by   Dan Alistarh, et al.
11

Motivated by applications to distributed optimization and machine learning, we consider the distributed mean estimation problem, in which n nodes are each assigned a multi-dimensional input vector, and must cooperate to estimate the mean of the input vectors, while minimizing communication. In this paper, we provide the first tight bounds for this problem, in terms of the trade-off between the amount of communication between nodes and the variance of the node estimates relative to the true value of the mean.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/14/2021

Leveraging Spatial and Temporal Correlations in Sparsified Mean Estimation

We study the problem of estimating at a central server the mean of a set...
research
01/12/2018

Communication Optimality Trade-offs For Distributed Estimation

This paper proposes Communication efficient REcursive Distributed estima...
research
06/24/2015

Communication Lower Bounds for Statistical Estimation Problems via a Distributed Data Processing Inequality

We study the tradeoff between the statistical error and communication co...
research
09/17/2022

Robust Online and Distributed Mean Estimation Under Adversarial Data Corruption

We study robust mean estimation in an online and distributed scenario in...
research
12/07/2021

Locally Differentially Private Sparse Vector Aggregation

Vector mean estimation is a central primitive in federated analytics. In...
research
11/24/2020

Wyner-Ziv Estimators: Efficient Distributed Mean Estimation with Side Information

Communication efficient distributed mean estimation is an important prim...
research
05/31/2022

Communication-efficient distributed eigenspace estimation with arbitrary node failures

We develop an eigenspace estimation algorithm for distributed environmen...

Please sign up or login with your details

Forgot password? Click here to reset