Reducing the computational and communication complexity of a distributed optimization for regularized logistic regression

Nozomi Miya, Hideyuki Masui, Hajime Jinushi, Toshiyasu Matsushima

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a new distributed optimization method that computes a Lasso estimator for logistic regression in the case when two parties have explanatory variables corresponding to distinct attributes. An existing protocol using the alternating direction method of multipliers (ADMM) for linear regression can be applied to logistic regression. However, this protocol needs an underlying iterative method such as the gradient method. We show that the proposed protocol using the generalized Bregman ADMM, which removes the necessity to use the underlying iterative method, requires lower computational and communication complexity.

Original languageEnglish
Title of host publication2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3454-3459
Number of pages6
ISBN (Electronic)9781728145693
DOIs
Publication statusPublished - 2019 Oct
Externally publishedYes
Event2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019 - Bari, Italy
Duration: 2019 Oct 62019 Oct 9

Publication series

NameConference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
Volume2019-October
ISSN (Print)1062-922X

Conference

Conference2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019
CountryItaly
CityBari
Period19/10/619/10/9

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Human-Computer Interaction

Fingerprint Dive into the research topics of 'Reducing the computational and communication complexity of a distributed optimization for regularized logistic regression'. Together they form a unique fingerprint.

Cite this