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

Many recent pattern recognition applications rely on complex distributedarchitectures in which sensing and computational nodes interact togetherthrough a communication network. Deep neural networks (DNNs) play an importantrole in this scenario, furnishing powerful decision mechanisms, at the price ofa high computational effort. Consequently, powerful state-of-the-art DNNs arefrequently split over various computational nodes, e.g., a first part stays onan embedded device and the rest on a server. Deciding where to split a DNN is achallenge in itself, making the design of deep learning applications even morecomplicated. Therefore, we propose Split-Et-Impera, a novel and practicalframework that i) determines the set of the best-split points of a neuralnetwork based on deep network interpretability principles without performing atedious try-and-test approach, ii) performs a communication-aware simulationfor the rapid evaluation of different neural network rearrangements, and iii)suggests the best match between the quality of service requirements of theapplication and the performance in terms of accuracy and latency time.

Split-Et-Impera: A Framework for the Design of Distributed Deep Learning Applications