Step-like dependence of memory function on pulse width in spintronics reservoir computing

TL;DR

This paper investigates how the pulse width of inputs affects the memory capacity in spintronics reservoir computing, revealing a step-like dependence linked to current-dependent magnetic relaxation dynamics.

Contribution

It uncovers the step-like relationship between input pulse width and memory capacity, highlighting the role of current-dependent magnetic damping in reservoir computing performance.

Findings

Memory capacity remains constant over a range of pulse widths.

Memory capacity drops sharply at longer pulse widths.

Analytical and numerical models explain the step-like behavior.

Abstract

Physical reservoir computing is a type of recurrent neural network that applies the dynamical response from physical systems to information processing. However, the relation between computation performance and physical parameters/phenomena still remains unclear. This study reports our progress regarding the role of current-dependent magnetic damping in the computational performance of reservoir computing. The current-dependent relaxation dynamics of a magnetic vortex core results in an asymmetric memory function with respect to binary inputs. A fast relaxation caused by a large input leads to a fast fading of the input memory, whereas a slow relaxation by a small input enables the reservoir to keep the input memory for a relatively long time. As a result, a step-like dependence is found for the short-term memory and parity-check capacities on the pulse width of input data, where the capacities remain at 1.5 for a certain range of the pulse width, and drop to 1.0 for a long pulse-width limit. Both analytical and numerical analyses clarify that the step-like behavior can be attributed to the current-dependent relaxation time of the vortex core to a limit-cycle state. }