Physical reservoir computing using magnetic skyrmion memristor and spin torque nano-oscillator
Physical reservoir computing using magnetic skyrmion memristor and spin torque nano-oscillator
复制标题
使用磁性斯格明子忆阻器和自旋扭矩纳米振荡器进行物理储层计算
DOI：
10.1063/1.5115183
复制
发表时间：
2019-11-04
影响因子：
4
通讯作者：
Liu, R. H.
中科院分区：
文献类型：
--
作者：
Jiang, Wencong;Chen, Lina;Liu, R. H.
自旋电子纳米器件具有纳秒级的超快非线性动力学和递归行为，有望实现高性能的自旋电子库计算（RC）系统。在这里，两个物理RC系统的基础上，一个单一的磁skyrmion忆阻器（MSM）和24个自旋扭矩纳米振荡器（STNO）的数值模拟处理图像分类任务和非线性动力学系统的预测，分别。基于MSM和STNO在电流脉冲刺激下的非线性响应，我们的结果表明，基于MSM的RC系统在图像分类方面表现出优异的性能，而基于STNO的RC系统在解决复杂的未知非线性动力学问题方面表现出色，例如，二阶非线性动力系统和NARMA10.我们的研究结果和分析的电流依赖的非线性动力学性质的MSM和STNO提供的策略，以优化实验参数，在构建更好的基于自旋电子学的类脑设备的机器学习为基础的computing.Spintronicnanodevices具有超快的非线性动力学和递归行为的纳秒尺度，有望使一个高性能的自旋电子水库计算（RC）系统。在这里，两个物理RC系统的基础上，一个单一的磁skyrmion忆阻器（MSM）和24个自旋扭矩纳米振荡器（STNO）的数值模拟处理图像分类任务和非线性动力学系统的预测，分别。基于MSM和STNO在电流脉冲刺激下的非线性响应，我们的结果表明，基于MSM的RC系统在图像分类方面表现出优异的性能，而基于STNO的RC系统在解决复杂的未知非线性动力学问题方面表现出色，例如，二阶非线性动力系统和NARMA10.我们的研究结果和对MSM和STNO的电流依赖性非线性动力学性质的分析为优化实验参数提供了策略，为构建更好的基于自旋电子学的类脑机器学习装置提供了依据。
Spintronic nanodevices have ultrafast nonlinear dynamic and recurrence behaviors on a nanosecond scale that promises to enable a high-performance spintronic reservoir computing (RC) system. Here, two physical RC systems based on one single magnetic skyrmion memristor (MSM) and 24 spin-torque nano-oscillators (STNOs) are numerically modeled to process image classification task and nonlinear dynamic system prediction, respectively. Based on the nonlinear responses of the MSM and STNO with current pulse stimulation, our results demonstrate that the MSM-based RC system exhibits excellent performance on image classification, while the STNO-based RC system does well in solving the complex unknown nonlinear dynamic problems, e.g., a second-order nonlinear dynamic system and NARMA10. Our result and analysis of the current-dependent nonlinear dynamic properties of the MSM and STNO provide the strategy to optimize the experimental parameters in building the better spintronic-based brainlike devices for machine learning based computing.Spintronic nanodevices have ultrafast nonlinear dynamic and recurrence behaviors on a nanosecond scale that promises to enable a high-performance spintronic reservoir computing (RC) system. Here, two physical RC systems based on one single magnetic skyrmion memristor (MSM) and 24 spin-torque nano-oscillators (STNOs) are numerically modeled to process image classification task and nonlinear dynamic system prediction, respectively. Based on the nonlinear responses of the MSM and STNO with current pulse stimulation, our results demonstrate that the MSM-based RC system exhibits excellent performance on image classification, while the STNO-based RC system does well in solving the complex unknown nonlinear dynamic problems, e.g., a second-order nonlinear dynamic system and NARMA10. Our result and analysis of the current-dependent nonlinear dynamic properties of the MSM and STNO provide the strategy to optimize the experimental parameters in building the better spintronic-based brainlike devices for machine lear...