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 success of deep neural networks (DNNs) is heavily dependent oncomputational resources. While DNNs are often employed on cloud servers, thereis a growing need to operate DNNs on edge devices. Edge devices are typicallylimited in their computational resources, yet, often multiple edge devices aredeployed in the same environment and can reliably communicate with each other.In this work we propose to facilitate the application of DNNs on the edge byallowing multiple users to collaborate during inference to improve theiraccuracy. Our mechanism, coined {\em edge ensembles}, is based on havingdiverse predictors at each device, which form an ensemble of models duringinference. To mitigate the communication overhead, the users share quantizedfeatures, and we propose a method for aggregating multiple decisions into asingle inference rule. We analyze the latency induced by edge ensembles,showing that its performance improvement comes at the cost of a minoradditional delay under common assumptions on the communication network. Ourexperiments demonstrate that collaborative inference via edge ensemblesequipped with compact DNNs substantially improves the accuracy over having eachuser infer locally, and can outperform using a single centralized DNN largerthan all the networks in the ensemble together.

Decentralized Low-Latency Collaborative Inference via Ensembles on the Edge