Adaptive Calibration: A Unified Conversion Framework of Spiking Neural Networks

arXiv:2311.14265v2 Announce Type: replace
Abstract: Spiking Neural Networks (SNNs) have emerged as a promising energy-efficient alternative to traditional Artificial Neural Networks (ANNs). Despite this, bridging the performance gap with ANNs in practical scenarios remains a significant challenge. This paper focuses on addressing the dual objectives of enhancing the performance and efficiency of SNNs through the established SNN Calibration conversion framework. Inspired by the biological nervous system, we propose a novel Adaptive-Firing Neuron Model (AdaFire) that dynamically adjusts firing patterns across …

abstract anns artificial artificial neural networks arxiv challenge conversion cs.cv energy framework gap networks neural networks paper performance practical spiking neural networks type