Online Asynchronous Distributed Regression

TL;DR

This paper introduces an asynchronous distributed regression method that leverages consensus algorithms for efficient nonparametric online learning, demonstrating strong performance and scalability on synthetic datasets with multiple processors.

Contribution

It proposes a novel consensus-based asynchronous distributed approach for online regression, extending distributed computation models to nonparametric learning tasks.

Findings

Excellent computation time and prediction accuracy on synthetic datasets.

Scalability demonstrated with up to 28 processors.

Asymptotic properties analyzed for the proposed method.

Abstract

Distributed computing offers a high degree of flexibility to accommodate modern learning constraints and the ever increasing size of datasets involved in massive data issues. Drawing inspiration from the theory of distributed computation models developed in the context of gradient-type optimization algorithms, we present a consensus-based asynchronous distributed approach for nonparametric online regression and analyze some of its asymptotic properties. Substantial numerical evidence involving up to 28 parallel processors is provided on synthetic datasets to assess the excellent performance of our method, both in terms of computation time and prediction accuracy.