A Probabilistic Synapse With Strained MTJs for Spiking Neural Networks
A Probabilistic Synapse With Strained MTJs for Spiking Neural Networks
复制标题
DOI：
10.1109/tnnls.2019.2917819
复制
发表时间：
2020-04
影响因子：
10.4
通讯作者：
S. Pagliarini;Sudipta Bhuin;Mehmet Meric Isgenc;A. Biswas;L. Pileggi
中科院分区：
文献类型：
--
作者：
S. Pagliarini;Sudipta Bhuin;Mehmet Meric Isgenc;A. Biswas;L. Pileggi
尖峰神经网络(SNN)对于那些传统计算受困于几乎无法克服的内存处理器瓶颈的应用程序很感兴趣。本文提出了一种随机SNN结构，它基于专门的记忆中逻辑突触单元来创建一个独特的处理系统，提供大规模的并行处理能力。我们建议的突触单元由应变磁隧道结(MTJ)设备和晶体管组成。我们突触中的MTJ是双重用途的，既用作随机位生成器，也用作通用存储器。我们的神经元被建模为具有阈值和折射的积分和激发组件。我们的电路采用与MTJ工艺兼容的CMOS28 nm工艺实现。我们的设计表明，建议的突触所需的面积仅为3.64美元~\MU\Text{m}^{2}/\Text{单位}$。空闲时，突触的功率为675 pw。当发射时，传播尖峰所需的能量为8.87 fJ。然后，我们演示了一个学习(无监督)MNIST数据库的手写数字并对其进行分类的SNN。模拟结果表明，即使在制造过程中存在可变性的情况下，我们的网络也表现出很高的分类效率。
Spiking neural networks (SNNs) are of interest for applications for which conventional computing suffers from the nearly insurmountable memory–processor bottleneck. This paper presents a stochastic SNN architecture that is based on specialized logic-in-memory synaptic units to create a unique processing system that offers massively parallel processing power. Our proposed synaptic unit consists of strained magnetic tunnel junction (MTJ) devices and transistors. MTJs in our synapse are dual purpose, used as both random bit generators and as general-purpose memory. Our neurons are modeled as integrate-and-fire components with thresholding and refraction. Our circuit is implemented using CMOS 28-nm technology that is compatible with the MTJ technology. Our design shows that the required area for the proposed synapse is only $3.64~\mu \text {m}^{2}/\text {unit}$ . When idle, the synapse consumes 675 pW. When firing, the energy required to propagate a spike is 8.87 fJ. We then demonstrate an SNN that learns (without supervision) and classifies handwritten digits of the MNIST database. Simulation results show that our network presents high classification efficiency even in the presence of fabrication variability.