CMOS-Free Multilayer Perceptron Enabled by Four-Terminal MTJ Device
CMOS-Free Multilayer Perceptron Enabled by Four-Terminal MTJ Device
复制标题
DOI：
--
复制
发表时间：
2020-02
期刊：
影响因子：
--
通讯作者：
Wesley H. Brigner;Naimul Hassan;Xuan Hu;C. Bennett;F. García-Sánchez;M. Marinella;J. Incorvia;J. Friedman
中科院分区：
文献类型：
--
作者：
Wesley H. Brigner;Naimul Hassan;Xuan Hu;C. Bennett;F. García-Sánchez;M. Marinella;J. Incorvia;J. Friedman
神经形态计算有望对处理非结构化信息的应用程序进行革命性的改进。为了充分实现这一潜力，神经形态系统应该利用新兴纳米器件的仿生行为。特别是，自旋电子器件的非易失性和模拟能力提供了特殊的机会。虽然先前已经提出了模拟神经元和突触的自旋电子器件，但需要互补金属氧化物半导体（CMOS）器件来实现多层自旋电子感知器交叉开关。因此，这项工作提出了一种新的自旋电子神经元，使纯粹的自旋电子多层感知器，消除了对CMOS电路的需要，简化了制造。
Neuromorphic computing promises revolutionary improvements over conventional systems for applications that process unstructured information. To fully realize this potential, neuromorphic systems should exploit the biomimetic behavior of emerging nanodevices. In particular, exceptional opportunities are provided by the non-volatility and analog capabilities of spintronic devices. While spintronic devices have previously been proposed that emulate neurons and synapses, complementary metal-oxide-semiconductor (CMOS) devices are required to implement multilayer spintronic perceptron crossbars. This work therefore proposes a new spintronic neuron that enables purely spintronic multilayer perceptrons, eliminating the need for CMOS circuitry and simplifying fabrication.