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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Spiking Neural Networks (SNNs) have recently emerged as the low-poweralternative to Artificial Neural Networks (ANNs) owing to their asynchronous,sparse, and binary information processing. To improve the energy-efficiency andthroughput, SNNs can be implemented on memristive crossbars whereMultiply-and-Accumulate (MAC) operations are realized in the analog domainusing emerging Non-Volatile-Memory (NVM) devices. Despite the compatibility ofSNNs with memristive crossbars, there is little attention to study on theeffect of intrinsic crossbar non-idealities and stochasticity on theperformance of SNNs. In this paper, we conduct a comprehensive analysis of therobustness of SNNs on non-ideal crossbars. We examine SNNs trained via learningalgorithms such as, surrogate gradient and ANN-SNN conversion. Our results showthat repetitive crossbar computations across multiple time-steps induce erroraccumulation, resulting in a huge performance drop during SNN inference. Wefurther show that SNNs trained with a smaller number of time-steps achievebetter accuracy when deployed on memristive crossbars.

Examining the Robustness of Spiking Neural Networks on Non-ideal Memristive Crossbars