SiTe CiM：面向超低精度深度神经网络的带符号三值存内计算
SiTe CiM: Signed Ternary Computing-in-Memory for Ultra-Low Precision Deep Neural Networks
AI总结：
提出支持带符号三值输入与权重点积的存内计算架构SiTe-CiM，基于交叉耦合位单元设计两种变体，在多种存储技术上实现显著降低延迟与能耗，并大幅提升系统吞吐量。
摘要：
三值深度神经网络（DNN）凭借其超低精度带来的低功耗运行优势，且仅有轻微的精度下降，在高度受限能耗系统中展现出巨大潜力。为在此类系统中实现高能效的硬件基础，我们提出了一种具备计算能力的存储器设计，称为SiTe-CiM，其特色在于支持带符号三值（SiTe）输入与权重之间点积的存内计算（CiM）。SiTe CiM基于两个位单元的交叉耦合，以实现带符号三值域中的点积CiM。我们基于8T-SRAM、3T嵌入式DRAM（3T-eDRAM）和3T铁电金属FET（FEMFET）存储器探索了SiTe CiM。我们提出了该技术的两种变体，即SiTe CiM I/II。在SiTe CiM I中，每个单元采用两个额外晶体管进行交叉耦合，实现了快速CiM操作，但会带来18%至34%的面积开销（与标准三值存储器相比）。在SiTe CiM II中，一列中每16个单元利用四个额外晶体管，因此仅产生6%的面积成本（但CiM速度比SiTe CiM I慢）。基于阵列分析，与各自的近存计算对应方案相比，在所考量的各种技术中，我们的设计实现了高达88%的CiM延迟降低和78%的CiM节能。此外，我们通过将SiTe CiM I/II阵列整合到三值DNN加速器中进行系统级分析，结果表明，与近存计算三值DNN加速器相比，吞吐量提升高达7倍，能耗降低高达2.5倍。
英文摘要：
Ternary Deep Neural Networks (DNN) have shown a large potential for highly energy-constrained systems by virtue of their low power operation (due to ultra-low precision) with only a mild degradation in accuracy. To enable an energy-efficient hardware substrate for such systems, we propose a compute-enabled memory design, referred to as SiTe-CiM, which features computing-in-memory (CiM) of dot products between signed ternary (SiTe) inputs and weights. SiTe CiM is based on cross-coupling of two bit cells to enable CiM of dot products in the signed ternary regime. We explore SiTe CiM with 8T-SRAM, 3T-embedded DRAM (3T-eDRAM) and 3T-ferroelectric metal FET (FEMFET) memories. We propose two flavors of this technique, namely SiTe CiM I/II. In SiTe CiM I, we employ two additional transistors per cell for cross-coupling, achieving fast CiM operations, albeit incurring an area overhead ranging from 18% to 34% (compared to standard ternary memories). In SiTe CiM II, four extra transistors are utilized for every 16 cells in a column, thereby incurring only 6% area cost (but leading to slower CiM than SiTe CiM I). Based on the array analysis, our designs achieve up to 88% lower CiM latency and 78% CiM energy savings across various technologies considered, as compared to their respective near-memory computing counterparts. Further, we perform system level analysis by incorporating SiTe CiM I/II arrays in a ternary DNN accelerator and show up to 7X throughput boost and up to 2.5X energy reduction compared to the near-memory ternary DNN accelerators.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
