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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In Federated edge learning (FEEL), energy-constrained devices at the networkedge consume significant energy when training and uploading their local machinelearning models, leading to a decrease in their lifetime. This work proposesnovel solutions for energy-efficient FEEL by jointly considering local trainingdata, available computation, and communications resources, and deadlineconstraints of FEEL rounds to reduce energy consumption. This paper considers asystem model where the edge server is equipped with multiple antennas employingbeamforming techniques to communicate with the local users through orthogonalchannels. Specifically, we consider a problem that aims to find the optimaluser's resources, including the fine-grained selection of relevant trainingsamples, bandwidth, transmission power, beamforming weights, and processingspeed with the goal of minimizing the total energy consumption given a deadlineconstraint on the communication rounds of FEEL. Then, we devise tractablesolutions by first proposing a novel fine-grained training algorithm thatexcludes less relevant training samples and effectively chooses only thesamples that improve the model's performance. After that, we derive closed-formsolutions, followed by a Golden-Section-based iterative algorithm to find theoptimal computation and communication resources that minimize energyconsumption. Experiments using MNIST and CIFAR-10 datasets demonstrate that ourproposed algorithms considerably outperform the state-of-the-art solutions asenergy consumption decreases by 79% for MNIST and 73% for CIFAR-10 datasets.

Fine-Grained Data Selection for Improved Energy Efficiency of Federated Edge Learning