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

Deep Neural Network has proved its potential in various perception tasks andhence become an appealing option for interpretation and data processing insecurity sensitive systems. However, security-sensitive systems demand not onlyhigh perception performance, but also design robustness under variouscircumstances. Unlike prior works that study network robustness from softwarelevel, we investigate from hardware perspective about the impact of SingleEvent Upset (SEU) induced parameter perturbation (SIPP) on neural networks. Wesystematically define the fault models of SEU and then provide the definitionof sensitivity to SIPP as the robustness measure for the network. We are thenable to analytically explore the weakness of a network and summarize the keyfindings for the impact of SIPP on different types of bits in a floating pointparameter, layer-wise robustness within the same network and impact of networkdepth. Based on those findings, we propose two remedy solutions to protect DNNsfrom SIPPs, which can mitigate accuracy degradation from 28% to 0.27% forResNet with merely 0.24-bit SRAM area overhead per parameter.

When Single Event Upset Meets Deep Neural Networks: Observations, Explorations, and Remedies