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

Gossip learning (GL), as a decentralized alternative to federated learning(FL), is more suitable for resource-constrained wireless networks, such asFANETs that are formed by unmanned aerial vehicles (UAVs). GL can significantlyenhance the efficiency and extend the battery life of UAV networks. Despite theadvantages, the performance of GL is strongly affected by data distribution,communication speed, and network connectivity. However, how these factorsinfluence the GL convergence is still unclear. Existing work studied theconvergence of GL based on a virtual quantity for the sake of convenience,which fail to reflect the real state of the network when some nodes areinaccessible. In this paper, we formulate and investigate the impact ofinaccessible nodes to GL under a dynamic network topology. We first decomposethe weight divergence by whether the node is accessible or not. Then, weinvestigate the GL convergence under the dynamic of node accessibility andtheoretically provide how the number of inaccessible nodes, datanon-i.i.d.-ness, and duration of inaccessibility affect the convergence.Extensive experiments are carried out in practical settings to comprehensivelyverify the correctness of our theoretical findings.

Technical Report: On the Convergence of Gossip Learning in the Presence of Node Inaccessibility