ACAS: Activation Consistency Coupled ANN-SNN framework for fast and memory-efficient SNN training

Jianxiong Tang, Jian-Huang Lai, Xiaohua Xie, Lingxiao Yang, Wei-Shi Zheng

Abstract: Highlights•Proposed the AC2AS framework for fast and memory efficient SNN training.•Designed an adaptive threshold adjustment algorithm and two spiking mapping units to ensure that the SNN can be equivalently represented by the ANN under the AC2AS framework.•The SNNs trained by AC2AS achieve considerable performance, sparse spike activity, and fast inference.