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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Unlike theoretical distributed learning (DL), DL over wireless edge networksfaces the inherent dynamics/uncertainty of wireless connections and edge nodes,making DL less efficient or even inapplicable under the highly dynamic wirelessedge networks (e.g., using mmW interfaces). This article addresses theseproblems by leveraging recent advances in coded computing and the deep duelingneural network architecture. By introducing coded structures/redundancy, adistributed learning task can be completed without waiting for stragglingnodes. Unlike conventional coded computing that only optimizes the codestructure, coded distributed learning over the wireless edge also requires tooptimize the selection/scheduling of wireless edge nodes with heterogeneousconnections, computing capability, and straggling effects. However, evenneglecting the aforementioned dynamics/uncertainty, the resulting jointoptimization of coding and scheduling to minimize the distributed learning timeturns out to be NP-hard. To tackle this and to account for the dynamics anduncertainty of wireless connections and edge nodes, we reformulate the problemas a Markov Decision Process and then design a novel deep reinforcementlearning algorithm that employs the deep dueling neural network architecture tofind the jointly optimal coding scheme and the best set of edge nodes fordifferent learning tasks without explicit information about the wirelessenvironment and edge nodes' straggling parameters. Simulations show that theproposed framework reduces the average learning delay in wireless edgecomputing up to 66% compared with other DL approaches. The jointly optimalframework in this article is also applicable to any distributed learning schemewith heterogeneous and uncertain computing nodes.

Joint Coding and Scheduling Optimization for Distributed Learning over Wireless Edge Networks