A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection

The Kaczmarz algorithm is an iterative method for solving systems of linear equations. We introduce a randomized Kaczmarz algorithm for solving systems of linear equations in a distributed environment, i.e., the equations within the system are distributed over multiple nodes within a network. The mo...

Full description

Bibliographic Details
Main Authors: Fritz Keinert, Eric S. Weber
Format: Article
Language:English
Published: MDPI AG 2022-02-01
Series:Axioms
Subjects:
Online Access:https://www.mdpi.com/2075-1680/11/3/106
_version_ 1797447190864461824
author Fritz Keinert
Eric S. Weber
author_facet Fritz Keinert
Eric S. Weber
author_sort Fritz Keinert
collection DOAJ
description The Kaczmarz algorithm is an iterative method for solving systems of linear equations. We introduce a randomized Kaczmarz algorithm for solving systems of linear equations in a distributed environment, i.e., the equations within the system are distributed over multiple nodes within a network. The modification we introduce is designed for a network with a tree structure that allows for passage of solution estimates between the nodes in the network. We demonstrate that the algorithm converges to the solution, or the solution of minimal norm, when the system is consistent. We also prove convergence rates of the randomized algorithm that depend on the spectral data of the coefficient matrix and the random control probability distribution. In addition, we demonstrate that the randomized algorithm can be used to identify anomalies in the system of equations when the measurements are perturbed by large, sparse noise.
first_indexed 2024-03-09T13:51:21Z
format Article
id doaj.art-a684dc8ea76b4bd084333175b83b6005
institution Directory Open Access Journal
issn 2075-1680
language English
last_indexed 2024-03-09T13:51:21Z
publishDate 2022-02-01
publisher MDPI AG
record_format Article
series Axioms
spelling doaj.art-a684dc8ea76b4bd084333175b83b60052023-11-30T20:49:58ZengMDPI AGAxioms2075-16802022-02-0111310610.3390/axioms11030106A Randomized Distributed Kaczmarz Algorithm and Anomaly DetectionFritz Keinert0Eric S. Weber1Department of Mathematics, Iowa State University, 396 Carver Hall, Ames, IA 50011, USADepartment of Mathematics, Iowa State University, 396 Carver Hall, Ames, IA 50011, USAThe Kaczmarz algorithm is an iterative method for solving systems of linear equations. We introduce a randomized Kaczmarz algorithm for solving systems of linear equations in a distributed environment, i.e., the equations within the system are distributed over multiple nodes within a network. The modification we introduce is designed for a network with a tree structure that allows for passage of solution estimates between the nodes in the network. We demonstrate that the algorithm converges to the solution, or the solution of minimal norm, when the system is consistent. We also prove convergence rates of the randomized algorithm that depend on the spectral data of the coefficient matrix and the random control probability distribution. In addition, we demonstrate that the randomized algorithm can be used to identify anomalies in the system of equations when the measurements are perturbed by large, sparse noise.https://www.mdpi.com/2075-1680/11/3/106Kaczmarz methodlinear equationsrandom controldistributed optimizationstochastic gradient descentsparse noise
spellingShingle Fritz Keinert
Eric S. Weber
A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection
Axioms
Kaczmarz method
linear equations
random control
distributed optimization
stochastic gradient descent
sparse noise
title A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection
title_full A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection
title_fullStr A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection
title_full_unstemmed A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection
title_short A Randomized Distributed Kaczmarz Algorithm and Anomaly Detection
title_sort randomized distributed kaczmarz algorithm and anomaly detection
topic Kaczmarz method
linear equations
random control
distributed optimization
stochastic gradient descent
sparse noise
url https://www.mdpi.com/2075-1680/11/3/106
work_keys_str_mv AT fritzkeinert arandomizeddistributedkaczmarzalgorithmandanomalydetection
AT ericsweber arandomizeddistributedkaczmarzalgorithmandanomalydetection
AT fritzkeinert randomizeddistributedkaczmarzalgorithmandanomalydetection
AT ericsweber randomizeddistributedkaczmarzalgorithmandanomalydetection