English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The conventional federated learning (FedL) architecture distributes machinelearning (ML) across worker devices by having them train local models that areperiodically aggregated by a server. FedL ignores two important characteristicsof contemporary wireless networks, however: (i) the network may containheterogeneous communication/computation resources, and (ii) there may besignificant overlaps in devices' local data distributions. In this work, wedevelop a novel optimization methodology that jointly accounts for thesefactors via intelligent device sampling complemented by device-to-device (D2D)offloading. Our optimization methodology aims to select the best combination ofsampled nodes and data offloading configuration to maximize FedL trainingaccuracy while minimizing data processing and D2D communication resourceconsumption subject to realistic constraints on the network topology and devicecapabilities. Theoretical analysis of the D2D offloading subproblem leads tonew FedL convergence bounds and an efficient sequential convex optimizer. Usingthese results, we develop a sampling methodology based on graph convolutionalnetworks (GCNs) which learns the relationship between network attributes,sampled nodes, and D2D data offloading to maximize FedL accuracy. Throughevaluation on popular datasets and real-world network measurements from ouredge testbed, we find that our methodology outperforms popular device samplingmethodologies from literature in terms of ML model performance, data processingoverhead, and energy consumption.

Device Sampling and Resource Optimization for Federated Learning in Cooperative Edge Networks