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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Spiking neural networks (SNNs), particularly the single-spike variant inwhich neurons spike at most once, are considerably more energy efficient thanstandard artificial neural networks (ANNs). However, single-spike SSNs aredifficult to train due to their dynamic and non-differentiable nature, wherecurrent solutions are either slow or suffer from training instabilities. Thesenetworks have also been critiqued for their limited computational applicabilitysuch as being unsuitable for time-series datasets. We propose a new model fortraining single-spike SNNs which mitigates the aforementioned training issuesand obtains competitive results across various image and neuromorphic datasets,with up to a $13.98\times$ training speedup and up to an $81\%$ reduction inspikes compared to the multi-spike SNN. Notably, our model performs on par withmulti-spike SNNs in challenging tasks involving neuromorphic time-seriesdatasets, demonstrating a broader computational role for single-spike SNNs thanpreviously believed.

Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks