Federated learning over wireless networks: A band-limited coordinated descent approach

Junshan Zhang, Na Li, Mehmet Dedeoglu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

We consider a many-to-one wireless architecture for federated learning at the network edge, where multiple edge devices collaboratively train a model using local data. The unreliable nature of wireless connectivity, together with constraints in computing resources at edge devices, dictates that the local updates at edge devices should be carefully crafted and compressed to match the wireless communication resources available and should work in concert with the receiver. Thus motivated, we propose SGD-based bandlimited coordinate descent algorithms for such settings. Specifically, for the wireless edge employing over-the-air computing, a common subset of k-coordinates of the gradient updates across edge devices are selected by the receiver in each iteration, and then transmitted simultaneously over k sub-carriers, each experiencing time-varying channel conditions. We characterize the impact of communication error and compression, in terms of the resulting gradient bias and mean squared error, on the convergence of the proposed algorithms. We then study learning-driven communication error minimization via joint optimization of power allocation and learning rates. Our findings reveal that optimal power allocation across different sub-carriers should take into account both the gradient values and channel conditions, thus generalizing the widely used water-filling policy. We also develop sub-optimal distributed solutions amenable to implementation.

Original languageEnglish (US)
Title of host publicationINFOCOM 2021 - IEEE Conference on Computer Communications
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780738112817
DOIs
StatePublished - May 10 2021
Event40th IEEE Conference on Computer Communications, INFOCOM 2021 - Vancouver, Canada
Duration: May 10 2021May 13 2021

Publication series

NameProceedings - IEEE INFOCOM
Volume2021-May
ISSN (Print)0743-166X

Conference

Conference40th IEEE Conference on Computer Communications, INFOCOM 2021
Country/TerritoryCanada
CityVancouver
Period5/10/215/13/21

ASJC Scopus subject areas

  • Computer Science(all)
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Federated learning over wireless networks: A band-limited coordinated descent approach'. Together they form a unique fingerprint.

Cite this