Literature Database Entry

wu2026stabilizing


Mengfan Wu, Mate Boban and Falko Dressler, "Stabilizing Asynchronous Decentralized Federated Learning via Representation Decoupling," IEEE Internet of Things Journal, 2026. (to appear)


Abstract

=Federated learning (FL) enables collaborative model training across distributed devices without sharing raw data. However, most existing FL frameworks rely on centralized coordination or strict synchronization, which limits their applicability in highly dynamic and decentralized Internet-of-Things environments. In fully asynchronous decentralized FL (DFL), exchanging complete models across heterogeneous devices can destabilize training, particularly under data imbalance, where device-specific classifiers diverge and disturb shared representation learning. In this work, we identify that the tight coupling between representation backbones and local classifiers is a key source of instability in asynchronous DFL. We propose a backbone–head decoupling framework that separates transferable feature extractors from device-specific prediction heads. Devices exchange only backbone parameters while training personalized heads locally. This structural separation mitigates cross-device interference from distribution-sensitive classifiers, stabilizes asynchronous updates, and also preserves personalization. Extensive experiments under diverse system dynamics and heterogeneous data distributions demonstrate that the proposed framework stabilizes asynchronous DFL, consistently improves personalization accuracy by over 10% in challenging scenarios, and maintains strong backbone generalization.

Quick access

BibTeX BibTeX

Contact

Mengfan Wu
Mate Boban
Falko Dressler

BibTeX reference

@article{wu2026stabilizing,
    author = {Wu, Mengfan and Boban, Mate and Dressler, Falko},
    note = {to appear},
    title = {{Stabilizing Asynchronous Decentralized Federated Learning via Representation Decoupling}},
    journal = {IEEE Internet of Things Journal},
    issn = {2327-4662},
    publisher = {IEEE},
    year = {2026},
   }
   
   

Copyright notice

Links to final or draft versions of papers are presented here to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted or distributed for commercial purposes without the explicit permission of the copyright holder.

The following applies to all papers listed above that have IEEE copyrights: Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

The following applies to all papers listed above that are in submission to IEEE conference/workshop proceedings or journals: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.

The following applies to all papers listed above that have ACM copyrights: ACM COPYRIGHT NOTICE. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept., ACM, Inc., fax +1 (212) 869-0481, or permissions@acm.org.

The following applies to all SpringerLink papers listed above that have Springer Science+Business Media copyrights: The original publication is available at www.springerlink.com.

This page was automatically generated using BibDB and bib2web.

Last modified: 2026-11-20