Stochastic Spiking Neural Networks Enabled by Magnetic Tunnel Junctions: From Nontelegraphic to Telegraphic Switching Regimes

TL;DR

This paper investigates the use of nanoelectronic spin devices, specifically magnetic tunnel junctions, for stochastic spiking neural networks, analyzing how different magnetic regimes affect system performance in neuromorphic computing tasks.

Contribution

It provides a detailed analysis of the impact of magnetic switching regimes on the design and performance of nanomagnet-based stochastic neural networks.

Findings

Low barrier magnets exhibit telegraphic switching behavior affecting network operation.

Network architecture must be adapted for superparamagnetic regime.

Device-to-system analysis demonstrates viability for digit recognition tasks.

Abstract

Stochastic spiking neural networks based on nanoelectronic spin devices can be a possible pathway to achieving "brainlike" compact and energy-effcient cognitive intelligence. The computational model attempt to exploit the intrinsic device stochasticity of nanoelectronic synaptic or neural components to perform learning or inference. However, there has been limited analysis on the scaling effect of stochastic spin devices and its impact on the operation of such stochastic networks at the system level. This work attempts to explore the design space and analyze the performance of nanomagnet-based stochastic neuromorphic computing architectures for magnets with different barrier heights. We illustrate how the underlying network architecture must be modified to account for the random telegraphic switching behavior displayed by magnets with low barrier heights as they are scaled into the superparamagnetic regime. We perform a device-to-system-level analysis on a deep neural-network architecture for a digit-recognition problem on the MNIST data set.