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

Federated Learning aims to train distributed deep models without sharing theraw data with the centralized server. Similarly, in distributed inference ofneural networks, by partitioning the network and distributing it across severalphysical nodes, activations and gradients are exchanged between physical nodes,rather than raw data. Nevertheless, when a neural network is partitioned anddistributed among physical nodes, failure of physical nodes causes the failureof the neural units that are placed on those nodes, which results in asignificant performance drop. Current approaches focus on resiliency oftraining in distributed neural networks. However, resiliency of inference indistributed neural networks is less explored. We introduce ResiliNet, a schemefor making inference in distributed neural networks resilient to physical nodefailures. ResiliNet combines two concepts to provide resiliency: skiphyperconnection, a concept for skipping nodes in distributed neural networkssimilar to skip connection in resnets, and a novel technique called failout,which is introduced in this paper. Failout simulates physical node failureconditions during training using dropout, and is specifically designed toimprove the resiliency of distributed neural networks. The results of theexperiments and ablation studies using three datasets confirm the ability ofResiliNet to provide inference resiliency for distributed neural networks.

ResiliNet: Failure-Resilient Inference in Distributed Neural Networks