APEX：用于精准高效SNN推理的双稀疏加速器
APEX: A Dual-Sparsity Accelerator for Precise and Efficient SNN Inference
AI总结：
APEX是集成PASC-IF神经元的双稀疏SNN推理加速器，可在低开销下实现比标准IF神经元更高的精度，同时降低能耗。
AI中文摘要：
脉冲神经网络（SNN）已成为人工神经网络（ANN）的节能替代方案，利用稀疏累加操作替代功耗高昂的乘累加操作。ANN-SNN转换是实现精度可与ANN媲美的深度SNN的广泛采用的方法。量化-裁剪-取整-移位（QCFS）激活可最小化转换误差，但需要大量推理时间步才能在真实视觉数据集上匹配源ANN的精度。PASCAL通过提出精准ANN-SNN转换积分-点火（PASC-IF）神经元解决了这一问题，该神经元保证转换后的SNN与源ANN的数学等价性，从而在显著减少时间步的情况下达到与ANN相当的精度。尽管有此算法进展，但部署PASC-IF神经元的硬件影响仍未被探索。在本研究中，我们提出APEX，一种双稀疏SNN推理加速器，将PASC-IF神经元集成到LoAS硬件框架中。三级PASC-IF数据通路被实现为完全组合电路，无额外延迟开销。APEX通过完全时间并行数据流利用输入脉冲和权重的双稀疏性，实现高效稀疏计算并减少内存流量。在所有评估模型中，PASC-IF神经元平均比标准IF神经元精度高3%，仅产生1.3%-5.4%的功耗开销、2.1%-2.7%的面积开销，在达到最佳精度的配置下能耗降低40%。
英文摘要：
Spiking Neural Networks (SNNs) have emerged as an energy-efficient alternative to Artificial Neural Networks (ANNs), leveraging sparse accumulate operations in the place of power-hungry multiply-and-accumulate operations. ANN-SNN conversion is a widely adopted approach to realize deep SNNs with accuracy comparable to that of ANNs. The Quantization-Clip-Floor-Shift (QCFS) activation minimizes conversion error, yet requires a large number of inference timesteps to match the source ANN accuracy on real-world vision datasets. PASCAL addresses this by proposing the Precise ANN-SNN Conversion Integrate-and-Fire (PASC-IF) neuron, which guarantees mathematical equivalence between the converted SNN and the source ANN, thereby achieving ANN-equivalent accuracy at significantly reduced timesteps. Despite this algorithmic advancement, the hardware implications of deploying the PASC-IF neuron remain unexplored. In this work, we present APEX, a dual-sparsity SNN inference accelerator that integrates the PASC-IF neuron into the LoAS hardware framework. The three-stage PASC-IF datapath is realized as a fully combinational circuit with no additional latency cost. APEX exploits dual sparsity in both input spikes and weights through a fully temporal-parallel dataflow, enabling efficient sparse computation and reduced memory traffic. Across all evaluated models, the PASC-IF neuron on average achieves up to 3% higher accuracy than the standard IF neuron, with a power overhead of only 1.3%-5.4%, an area overhead of 2.1%-2.7%, and 40% energy reduction for best accuracy configurations.
