A compact, accurate, and bitwidth-programmable in-memory computing (IMC) static random-access memory (SRAM) macro, named CAP-RAM, is presented for energy-efficient convolutional neural network (CNN) inference. It leverages a novel charge-domain multiply-and-accumulate (MAC) mechanism and circuitry to achieve superior linearity under process variations compared to conventional IMC designs. The adopted semi-parallel architecture efficiently stores filters from multiple CNN layers by sharing eight standard 6T SRAM cells with one charge-domain MAC circuit. Moreover, up to six levels of bit-width of weights with two encoding schemes and eight levels of input activations are supported. A 7-bit charge-injection SAR (ciSAR) analog-to-digital converter (ADC) getting rid of sample and hold (S&H) and input/reference buffers further improves the overall energy efficiency and throughput. A 65-nm prototype validates the excellent linearity and computing accuracy of CAP-RAM. A single 512x128 macro stores a complete pruned and quantized CNN model to achieve 98.8% inference accuracy on the MNIST data set and 89.0% on the CIFAR-10 data set, with a 573.4-giga operations per second (GOPS) peak throughput and a 49.4-tera operations per second (TOPS)/W energy efficiency.
翻译:在节能进化神经网络(CNN)的推论中,介绍了一个叫CAP-RAM(CAP-RAM)的精密、精确和比特维的静态随机存取存储器(IMC)(静态随机存取存储器)宏,称为CAP-RAM(SRAM),用于高能效的进化神经网络(CNN)的推论。它利用一种新型的充电多元和累积(MAC)机制和电路,在与常规的IMC设计相比的流程变异下实现超强线性。所采用的半平行结构通过共享8个标准的6T SRAM(静态的)静态随机存存储器(SRAM),一个充电多管回路路。此外,还支持高达6个位比维重的重量比特(Bitwithwi-wima-cal),一个512x的IMIS-RAM(IS-RAM)运行第518次的精确度,一个IMIS-10的IMSA(IS-10)完整的IMIS-O(S-I-I-I)运行第98-10号,一个完整的SIMIS-IMIS-I-10号运行,一个98-IMIS-I-IMIS-I-10号的完整数据全的精确的528(S-IMIS-IMIS-I-I-I-I-I-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-Q-PQ-PQ),一个完整的精确度,一个完整的精确度数据数据集,一个完整的完整的运行,一个完整的精确度,一个完整的运行到一个全的512-Q-PIAS-Q-Q-PI-Q-Q-Q-Q-Q-PQ-P-Q-Q-Q-P-Q-I-Q-Q-Q-Q-P-P-IFAS-IAS-I-IAS-IAS-IAS-I-I-8个运行,一个完整的完整的完整的完整的完整的完整的完整的运行的精确的精确的精确的运行的运行的运行的精确的精确的精确的运行的运行的运行的固定的固定的运行的固定的