University of Oulu

T. AlShammari, S. Samarakoon, A. Elgabli and M. Bennis, "BayGo: Joint Bayesian Learning and Information-Aware Graph Optimization," ICC 2021 - IEEE International Conference on Communications, 2021, pp. 1-6, doi: 10.1109/ICC42927.2021.9500613

BayGo : joint Bayesian learning and information-aware graph optimization

Saved in:
Author: AlShammari, Tamara1; Samarakoon, Sumudu1; Elgabli, Anis1;
Organizations: 1Centre for Wireless Communication, University of Oulu, Finland
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 0.8 MB)
Persistent link: http://urn.fi/urn:nbn:fi-fe2021102151850
Language: English
Published: Institute of Electrical and Electronics Engineers, 2021
Publish Date: 2021-10-21
Description:

Abstract

This article deals with the problem of distributed machine learning, in which agents update their models based on their local datasets, and aggregate the updated models collaboratively and in a fully decentralized manner. In this paper, we tackle the problem of information heterogeneity arising in multi-agent networks where the placement of informative agents plays a crucial role in the learning dynamics. Specifically, we propose BayGo, a novel fully decentralized joint Bayesian learning and graph optimization framework with proven fast convergence over a sparse graph. Under our framework, agents are able to learn and communicate with the most informative agent to their own learning. Unlike prior works, our framework assumes no prior knowledge of the data distribution across agents nor does it assume any knowledge of the true parameter of the system. The proposed alternating minimization based framework ensures global connectivity in a fully decentralized way while minimizing the number of communication links. We theoretically show that by optimizing the proposed objective function, the estimation error of the posterior probability distribution decreases exponentially at each iteration. Via extensive simulations, we show that our framework achieves faster convergence and higher accuracy compared to fully-connected and star topology graphs.

see all

Series: IEEE International Conference on Communications
ISSN: 1550-3607
ISSN-E: 1938-1883
ISSN-L: 1550-3607
ISBN: 978-1-7281-7122-7
ISBN Print: 978-1-7281-7123-4
Pages: 1 - 6
DOI: 10.1109/ICC42927.2021.9500613
OADOI: https://oadoi.org/10.1109/ICC42927.2021.9500613
Host publication: ICC 2021 - IEEE International Conference on Communications
Conference: IEEE International Conference on Communications
Type of Publication: A4 Article in conference proceedings
Field of Science: 213 Electronic, automation and communications engineering, electronics
Subjects:
Funding: This work was supported in part by the INFOTECH Project NOOR, in part by the NEGEIN project, by the EU-CHISTERA projects LeadingEdge and CONNECT, the EU-H2020 project under agreement No. 957218 (IntellIoT) and the Academy of Finland projects MISSION and SMARTER.
EU Grant Number: (957218) IntellIoT - Intelligent, distributed, human-centered and trustworthy IoT environments
Copyright information: © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.