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 popular federated edge learning (FEEL) framework allowsprivacy-preserving collaborative model training via frequent learning-updatesexchange between edge devices and server. Due to the constrained bandwidth,only a subset of devices can upload their updates at each communication round.This has led to an active research area in FEEL studying the optimal devicescheduling policy for minimizing communication time. However, owing to thedifficulty in quantifying the exact communication time, prior work in this areacan only tackle the problem partially by considering either the communicationrounds or per-round latency, while the total communication time is determinedby both metrics. To close this gap, we make the first attempt in this paper toformulate and solve the communication time minimization problem. We firstderive a tight bound to approximate the communication time throughcross-disciplinary effort involving both learning theory for convergenceanalysis and communication theory for per-round latency analysis. Building onthe analytical result, an optimized probabilistic scheduling policy is derivedin closed-form by solving the approximate communication time minimizationproblem. It is found that the optimized policy gradually turns its priorityfrom suppressing the remaining communication rounds to reducing per-roundlatency as the training process evolves. The effectiveness of the proposedscheme is demonstrated via a use case on collaborative 3D objective detectionin autonomous driving.

Accelerating Federated Edge Learning via Optimized Probabilistic Device Scheduling