Spiking neural networks (SNNs) that enable low-power design on edge devices have recently attracted significant research. However, the temporal characteristic of SNNs causes high latency, high bandwidth and high energy consumption for the hardware. In this work, we propose a binary weight spiking model with IF-based Batch Normalization for small time steps and low hardware cost when direct training with input encoding layer and spatio-temporal back propagation (STBP). In addition, we propose a vectorwise hardware accelerator that is reconfigurable for different models, inference time steps and even supports the encoding layer to receive multi-bit input. The required memory bandwidth is further reduced by two-layer fusion mechanism. The implementation result shows competitive accuracy on the MNIST and CIFAR-10 datasets with only 8 time steps, and achieves power efficiency of 25.9 TOPS/W.
翻译:在这项工作中,我们建议采用基于IF的批量标准化二重制模式,用于小时间步骤和低硬件成本,如果直接培训输入编码层和时空反波传播(STBP),则采用二分制重制模型。此外,我们提议采用一种矢量式硬件加速器,可以对不同的模型、推导时间步骤、甚至支持编码层进行重新配置,以接收多位输入。所需的内存带宽被两层聚合机制进一步缩小。实施结果显示,MNIST和CIFAR-10数据集的竞争性精度只有8个步骤,并实现了25.9 TOPS/W的功率效率。