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

Due to the binary spike signals making converting the traditional high-powermultiply-accumulation (MAC) into a low-power accumulation (AC) available, thebrain-inspired Spiking Neural Networks (SNNs) are gaining more and moreattention. However, the binary spike propagation of the Full-Spike NeuralNetworks (FSNN) with limited time steps is prone to significant informationloss. To improve performance, several state-of-the-art SNN models trained fromscratch inevitably bring many non-spike operations. The non-spike operationscause additional computational consumption and may not be deployed on someneuromorphic hardware where only spike operation is allowed. To train alarge-scale FSNN with high performance, this paper proposes a novel Dual-StreamTraining (DST) method which adds a detachable Auxiliary Accumulation Pathway(AAP) to the full spiking residual networks. The accumulation in AAP couldcompensate for the information loss during the forward and backward of fullspike propagation, and facilitate the training of the FSNN. In the test phase,the AAP could be removed and only the FSNN remained. This not only keeps thelower energy consumption but also makes our model easy to deploy. Moreover, forsome cases where the non-spike operations are available, the APP could also beretained in test inference and improve feature discrimination by introducing alittle non-spike consumption. Extensive experiments on ImageNet, DVS Gesture,and CIFAR10-DVS datasets demonstrate the effectiveness of DST.

Training Full Spike Neural Networks via Auxiliary Accumulation Pathway