Literature Database Entry

tao2026autosub

Youming Tao, Meryem Simsek and Falko Dressler, "AutoSub: Autonomous Submodel Averaging," Proceedings of IEEE Annual Congress on Artificial Intelligence of Things (AioT 2026), Kobe, Japan, December 2026. (to appear)

Abstract

Federated learning on edge devices must accommodate not only device availability, but also limits on model capacity and local computation. These limits can change over time, and devices can observe their current resource conditions locally. We study autonomous federated learning (AutoFL), where devices choose when to participate, how much of the model to train, and how much local work to perform. We develop AutoSub (Autonomous Submodel Averaging) to combine these contributions into one shared model. Each participating device uniformly samples a structured submodel of the size it has chosen and keeps it fixed throughout local training. It returns the average of its local gradient directions, normalizing for the local step count and stepsize. The server aggregates updates whenever its buffer contains the prescribed number of messages. We evaluate AutoSub on CIFAR-10 and CIFAR-100 in a synthetic event-driven simulation with changing memory budgets, local training-time budgets, and link rates. In these experiments, AutoSub achieves higher mean test accuracy than the baselines at the final evaluation point, after the same number of server updates, and can train at a larger fraction of participation opportunities. It also outperforms the variant without workload normalization. The accuracy gap increases in an experiment that moves from equal local step counts to unequal local workloads by changing training-time budgets and device speeds.

Quick access

BibTeX BibTeX

Contact

Youming Tao
Meryem Simsek
Falko Dressler

BibTeX reference

@inproceedings{tao2026autosub,
    author = {Tao, Youming and Simsek, Meryem and Dressler, Falko},
    note = {to appear},
    title = {{AutoSub: Autonomous Submodel Averaging}},
    publisher = {IEEE},
    address = {Kobe, Japan},
    booktitle = {IEEE Annual Congress on Artificial Intelligence of Things (AioT 2026)},
    month = {12},
    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: 2026-10-27