arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.02193physics.app-phcond-mat.mtrl-sciphysics.comp-ph
收藏

波速和大小之间的通用缩放使得基于传播自旋波的纳米级高性能储层计算成为可能

Universal scaling between wave speed and size enables nanoscale high-performance reservoir computing based on propagating spin-waves

Satoshi Iihama, Yuya Koike, Shigemi Mizukami, Natsuhiko Yoshinaga

摘要:

利用自旋波的神经形态计算对于高速纳米器件是很有前途的,但是高性能的实现还没有实现。在这里,我们表明,使用微磁模拟和响应函数的简化理论,自旋波物理水库计算可以实现小型化到纳米尺度保持高计算能力与其他国家的最先进的系统。我们还表明,系统尺寸的缩放与自旋波的传播速度起着关键作用,以实现在纳米尺度上的高性能。

英文摘要:

Neuromorphic computing using spin waves is promising for high-speed nanoscale devices, but the realization of high performance has not yet been achieved. Here we show, using micromagnetic simulations and simplified theory with response functions, that spin-wave physical reservoir computing can achieve miniaturization down to nanoscales keeping high computational power comparable with other state-of-art systems. We also show the scaling of system sizes with the propagation speed of spin waves plays a key role to achieve high performance at nanoscales.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑