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

Asynchronous federated learning aims to solve the straggler problem inheterogeneous environments, i.e., clients have small computational capacitiesthat could cause aggregation delay. The principle of asynchronous federatedlearning is to allow the server to aggregate the model once it receives anupdate from any client rather than waiting for updates from multiple clients orwaiting a specified amount of time in the synchronous mode. Due to theasynchronous setting, the stale model problem could occur, where the slowclients could utilize an outdated local model for their local data training.Consequently, when these locally trained models are uploaded to the server,they may impede the convergence of the global training. Therefore, effectivemodel aggregation strategies play a significant role in updating the globalmodel. Besides, client scheduling is also critical when heterogeneous clientswith diversified computing capacities are participating in the federatedlearning process. This work first investigates the impact of the convergence ofasynchronous federated learning mode when adopting the aggregation coefficientin synchronous mode. The effective aggregation solutions that can achieve thesame convergence result as in the synchronous mode are then proposed, followedby an improved aggregation method with client scheduling. The simulationresults in various scenarios demonstrate that the proposed algorithm convergeswith a similar level of accuracy as the classical synchronous federatedlearning algorithm but effectively accelerates the learning process, especiallyin its early stage.

CSMAAFL: Client Scheduling and Model Aggregation in Asynchronous Federated Learning