Initialisation and network effects in decentralised federated learning

Thumbnail Image

Files

Initialisation_and_network_effects_in_decentralised_federated_learning.pdf (2.99 MB)

Access rights

openAccess
CC BY
publishedVersion

URL

Journal Title

Journal ISSN

Volume Title

A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2025-12

Department

Department of Computer Science

Major/Subject

Mcode

Degree programme

Language

en

Pages

22

Series

Applied Network Science, Volume 10, issue 1, pp. 1-22

Abstract

Fully decentralised federated learning enables collaborative training of individual machine learning models on a distributed network of communicating devices while keeping the training data localised on each node. This approach avoids central coordination, enhances data privacy and eliminates the risk of a single point of failure. Our research highlights that the effectiveness of decentralised federated learning is significantly influenced by the network topology of connected devices and the initial conditions of the learning models. We propose a strategy for uncoordinated initialisation of the artificial neural networks based on the distribution of eigenvector centralities of the underlying communication network, leading to a radically improved training efficiency. Additionally, our study explores the scaling behaviour and the choice of environmental parameters under our proposed initialisation strategy. This work paves the way for more efficient and scalable artificial neural network training in a distributed and uncoordinated environment, offering a deeper understanding of the intertwining roles of network structure and learning dynamics.

Description

Publisher Copyright: © The Author(s) 2025.

Keywords

Complex networks, Federated learning, Gossip protocols, Random walks

Other note

DOI

10.1007/s41109-025-00737-4

Citation

Badie-Modiri, A, Boldrini, C, Valerio, L, Kertész, J & Karsai, M 2025, 'Initialisation and network effects in decentralised federated learning', Applied Network Science, vol. 10, no. 1, 53, pp. 1-22. https://doi.org/10.1007/s41109-025-00737-4

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202511128753

Collections

[article-cris] Perustieteiden korkeakoulu / SCI