Online Model Updating with Analog Aggregation in Wireless Edge Learning

Juncheng Wang, Min Dong, Ben Liang, Gary Boudreau, Hatem Abou-Zeid

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

5 Citations (Scopus)

Abstract

We consider federated learning in a wireless edge network, where multiple power-limited mobile devices collaboratively train a global model, using their local data with the assistance of an edge server. Exploiting over-the-air computation, the edge server updates the global model via analog aggregation of the local models over noisy wireless fading channels. Unlike existing works that separately optimize computation and communication at each step of the learning algorithm, in this work, we jointly optimize the training of the global model and the analog aggregation of local models over time. Our objective is to minimize the accumulated training loss at the edge server, subject to individual long-term transmit power constraints at the mobile devices. We propose an efficient algorithm, termed Online Model Updating with Analog Aggregation (OMUAA), to adaptively update the local and global models based on the time-varying communication environment. The trained model of OMUAA is channel- and power-aware, and it is in closed form with low computational complexity. We study the mutual impact between model training and analog aggregation over time, to derive performance bounds on the computation and communication performance metrics. Simulation results based on real-world image classification datasets and typical Long-Term Evolution network settings demonstrate substantial performance gain of OMUAA over the known best alternatives.
Original languageEnglish
Title of host publicationINFOCOM 2022 - IEEE Conference on Computer Communications
PublisherIEEE
Pages1229-1238
Number of pages10
ISBN (Electronic)9781665458221
ISBN (Print)9781665458238
DOIs
Publication statusPublished - May 2022
Event41st IEEE Conference on Computer Communications, IEEE INFOCOM 2022 - Online
Duration: 2 May 20225 May 2022
https://infocom2022.ieee-infocom.org/index.html
https://ieeexplore.ieee.org/xpl/conhome/9796607/proceeding

Publication series

NameProceedings of IEEE Conference on Computer Communications
PublisherIEEE
Volume2022-May
ISSN (Print)0743-166X
ISSN (Electronic)2641-9874

Conference

Conference41st IEEE Conference on Computer Communications, IEEE INFOCOM 2022
CityOnline
Period2/05/225/05/22
Internet address

User-Defined Keywords

  • Federated learning
  • Wireless edge networks
  • Over-the-air computation
  • Analog aggregation
  • Online optimization
  • Long-term constraints

Fingerprint

Dive into the research topics of 'Online Model Updating with Analog Aggregation in Wireless Edge Learning'. Together they form a unique fingerprint.

Cite this