Transformer inference requires high compute accuracy; achieving this using analog CIMs has been difficult due to inherent computational errors. To overcome this challenge, we propose a Capacitor-Reconfiguring CIM (CR-CIM) to realize high compute accuracy analog CIM with a 10-bit ADC attaining high-area/power efficiency. CR-CIM reconfigures its capacitor array to serve dual purposes: for computation and ADC conversion, achieving significant area savings. Furthermore, CR-CIMs eliminate signal attenuation by keeping the signal charge stationary during operation, leading to a 4x improvement in comparator energy efficiency. We also propose a software-analog co-design technique integrating majority voting into the 10-bit ADC to dynamically optimize the CIM noise performance based on the running layer to further save inference power. Our CR-CIM achieves the highest compute-accuracy for analog CIMs, and the power efficiency of 818 TOPS/W is competitive with the state-of-the-art. Furthermore, the FoM considering SQNR and CSNR is 2.3x and 1.5x better than previous works, respectively. Vision Transformer (ViT) inference is achieved and realizes a highest CIFAR10 accuracy of 95.8% for analog CIMs.
翻译:变换器的推断要求高计算精度; 使用模拟 CIM 实现此功能由于内在的计算错误而困难重重。 为了克服这一挑战,我们建议使用CIM (CR-CIM) 能力再配置 CIM (CR-CIM) 实现高计算精度模拟 CIM, 其10位ADC 达到高面积/电能效率。 CR- CIM 重新配置其电容器阵列, 以达到双重目的: 用于计算和 ADC 转换, 实现大量地区节约。 此外, CR- CIM 保持信号电荷固定状态, 从而消除信号衰减, 导致参照器能源效率4x 的提高。 我们还建议采用软件- 软件- 配置共同设计技术, 将多数投票纳入 10位ADC (CR- CIM), 以动态优化 CIM 音效, 以进一步节省电能。 CRIM 达到最高级的计算精确度, 并且 818 TOPS/W 的电量效率与状态相比具有竞争力。 此外, FOM 考虑 SQNRN 和 CS 10 的CIM 和 CRVI 分别实现了实现了 和1.5x 最高精确度。