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

The valuable data collected by IoT devices in edge networks together with theresurgence of ML stimulate the latest trend of edge AI. However, recent FLmethods face major challenges including communication bottleneck, dataheterogeneity and security concerns in edge IoT scenarios, especially whenbeing adopted for distributed learning among massive IoT devices equipped withlimited data and transmission resources. Meanwhile, the swarm nature of IoTsystems is overlooked by most existing literature, which calls for new designsof distributed learning algorithms. Inspired by the success of biologicalintelligence (BI) of gregarious organisms, we propose a novel edge learningapproach for swarm IoT, called communication-efficient and Byzantine-robustdistributed swarm learning (CB-DSL), through a holistic integration ofAI-enabled stochastic gradient descent and BI-enabled particle swarmoptimization. To deal with non-i.i.d. data issues and Byzantine attacks, globaldata samples are introduced in CB-DSL and shared among IoT workers, which notonly alleviates the local data heterogeneity effectively but also enables tofully utilize the exploration-exploitation mechanism of swarm intelligence.Further, we provide convergence analysis to theoretically demonstrate that theproposed CB-DSL is superior to the standard FL with better convergencebehavior. In addition, to measure the effectiveness of the introduction of theglobally shared dataset, we also evaluate the model divergence by deriving itsupper bound, which is related to the distance between the data distribution atlocal IoT devices and the population distribution for the whole datasets.Numerical results verify that the proposed CB-DSL outperforms the existingbenchmarks in terms of faster convergence speed, higher convergent accuracy,lower communication cost, and better robustness against non-i.i.d. data andByzantine attacks.

CB-DSL: Communication-efficient and Byzantine-robust Distributed Swarm Learning on Non-i.i.d. Data