Spiking Neural Networks (SNNs) have shown capabilities of achieving high accuracy under unsupervised settings and low operational power/energy due to their bio-plausible computations. Previous studies identified that DRAM-based off-chip memory accesses dominate the energy consumption of SNN processing. However, state-of-the-art works do not optimize the DRAM energy-per-access, thereby hindering the SNN-based systems from achieving further energy efficiency gains. To substantially reduce the DRAM energy-per-access, an effective solution is to decrease the DRAM supply voltage, but it may lead to errors in DRAM cells (i.e., so-called approximate DRAM). Towards this, we propose \textit{EnforceSNN}, a novel design framework that provides a solution for resilient and energy-efficient SNN inference using reduced-voltage DRAM for embedded systems. The key mechanisms of our EnforceSNN are: (1) employing quantized weights to reduce the DRAM access energy; (2) devising an efficient DRAM mapping policy to minimize the DRAM energy-per-access; (3) analyzing the SNN error tolerance to understand its accuracy profile considering different bit error rate (BER) values; (4) leveraging the information for developing an efficient fault-aware training (FAT) that considers different BER values and bit error locations in DRAM to improve the SNN error tolerance; and (5) developing an algorithm to select the SNN model that offers good trade-offs among accuracy, memory, and energy consumption. The experimental results show that our EnforceSNN maintains the accuracy (i.e., no accuracy loss for BER less-or-equal 10^-3) as compared to the baseline SNN with accurate DRAM, while achieving up to 84.9\% of DRAM energy saving and up to 4.1x speed-up of DRAM data throughput across different network sizes.

翻译：脉冲神经网络（SNN）因其类脑计算特性，在无监督条件下展现出高精度与低功耗/能量优势。已有研究表明，基于DRAM的片外存储器访问是SNN处理能耗的主要来源。然而，现有技术并未优化DRAM单位访问能耗，限制了SNN系统能效的进一步提升。为显著降低DRAM单位访问能耗，一种有效方案是降低DRAM供电电压，但这可能导致DRAM单元产生错误（即所谓近似DRAM）。为此，我们提出\textit{EnforceSNN}——一种面向嵌入式系统采用低压DRAM实现鲁棒节能SNN推理的创新设计框架。其核心机制包括：（1）采用量化权重降低DRAM访问能耗；（2）设计高效DRAM映射策略最小化单位访问能耗；（3）分析SNN容错特性以理解不同误比特率（BER）下的精度分布；（4）利用该信息开发考虑DRAM不同BER值与错误比特位置的高效故障感知训练（FAT）方法，提升SNN容错能力；（5）构建算法选择可在精度、存储与能耗间取得最佳平衡的SNN模型。实验结果表明，与采用精确DRAM的基准SNN相比，EnforceSNN在维持精度（BER≤10^{-3}时无精度损失）的同时，可实现高达84.9%的DRAM节能效果，并在不同网络规模下获得最高4.1倍的DRAM数据吞吐量加速。