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

The emergence of brain-inspired neuromorphic computing as a paradigm for edgeAI is motivating the search for high-performance and efficient spiking neuralnetworks to run on this hardware. However, compared to classical neuralnetworks in deep learning, current spiking neural networks lack competitiveperformance in compelling areas. Here, for sequential and streaming tasks, wedemonstrate how a novel type of adaptive spiking recurrent neural network(SRNN) is able to achieve state-of-the-art performance compared to otherspiking neural networks and almost reach or exceed the performance of classicalrecurrent neural networks (RNNs) while exhibiting sparse activity. From this,we calculate a $>$100x energy improvement for our SRNNs over classical RNNs onthe harder tasks. To achieve this, we model standard and adaptivemultiple-timescale spiking neurons as self-recurrent neural units, and leveragesurrogate gradients and auto-differentiation in the PyTorch Deep Learningframework to efficiently implement backpropagation-through-time, includinglearning of the important spiking neuron parameters to adapt our spikingneurons to the tasks.

Effective and Efficient Computation with Multiple-timescale Spiking Recurrent Neural Networks