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

Federated learning allows mobile devices, i.e., workers, to use their localdata to collaboratively train a global model required by the model owner.Federated learning thus addresses the privacy issues of traditional machinelearning. However, federated learning faces the energy constraints of theworkers and the high network resource cost due to the fact that a number ofglobal model transmissions may be required to achieve the target accuracy. Toaddress the energy constraint, a power beacon can be used that recharges energyto the workers. However, the model owner may need to pay an energy cost to thepower beacon for the energy recharge. To address the high network resourcecost, the model owner can use a WiFi channel, called default channel, for theglobal model transmissions. However, communication interruptions may occur dueto the instability of the default channel quality. For this, special channelssuch as LTE channels can be used, but this incurs channel cost. As such, theproblem of the model owner is to decide amounts of energy recharged to theworkers and to choose channels used to transmit its global model to the workersto maximize the number of global model transmissions while minimizing theenergy and channel costs. This is challenging for the model owner under theuncertainty of the channel, energy and mobility states of the workers. In thispaper, we thus propose to employ the Deep Q-Network (DQN) that enables themodel owner to find the optimal decisions on the energy and the channelswithout any a priori network knowledge. Simulation results show that theproposed DQN always achieves better performance compared to the conventionalalgorithms.

Resource Allocation in Mobility-Aware Federated Learning Networks: A Deep Reinforcement Learning Approach