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

Spiking Neural Networks (SNNs), as bio-inspired energy-efficient neuralnetworks, have attracted great attentions from researchers and industry. Themost efficient way to train deep SNNs is through ANN-SNN conversion. However,the conversion usually suffers from accuracy loss and long inference time,which impede the practical application of SNN. In this paper, we theoreticallyanalyze ANN-SNN conversion and derive sufficient conditions of the optimalconversion. To better correlate ANN-SNN and get greater accuracy, we proposeRate Norm Layer to replace the ReLU activation function in source ANN training,enabling direct conversion from a trained ANN to an SNN. Moreover, we proposean optimal fit curve to quantify the fit between the activation value of sourceANN and the actual firing rate of target SNN. We show that the inference timecan be reduced by optimizing the upper bound of the fit curve in the revisedANN to achieve fast inference. Our theory can explain the existing work on fastreasoning and get better results. The experimental results show that theproposed method achieves near loss less conversion with VGG-16,PreActResNet-18, and deeper structures. Moreover, it can reach 8.6x fasterreasoning performance under 0.265x energy consumption of the typical method.The code is available athttps://github.com/DingJianhao/OptSNNConvertion-RNL-RIL.

Optimal ANN-SNN Conversion for Fast and Accurate Inference in Deep Spiking Neural Networks