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

Recent research demonstrated the promise of using resistive random accessmemory (ReRAM) as an emerging technology to perform inherently parallel analogdomain in-situ matrix-vector multiplication -- the intensive and keycomputation in deep neural networks (DNNs). However, hardware failure, such asstuck-at-fault defects, is one of the main concerns that impedes the ReRAMdevices to be a feasible solution for real implementations. The existingsolutions to address this issue usually require an optimization to be conductedfor each individual device, which is impractical for mass-produced products(e.g., IoT devices). In this paper, we rethink the value of weight pruning inReRAM-based DNN design from the perspective of model fault tolerance. And adifferential mapping scheme is proposed to improve the fault tolerance under ahigh stuck-on fault rate. Our method can tolerate almost an order of magnitudehigher failure rate than the traditional two-column method in representativeDNN tasks. More importantly, our method does not require extra hardware costcompared to the traditional two-column mapping scheme. The improvement isuniversal and does not require the optimization process for each individualdevice.

Improving DNN Fault Tolerance using Weight Pruning and Differential Crossbar Mapping for ReRAM-based Edge AI