Shape-based Magnetic Domain Wall Drift for an Artificial Spintronic Leaky Integrate-and-Fire Neuron

TL;DR

This paper introduces a shape-engineered magnetic tunnel junction neuron that intrinsically performs the leaking function of a leaky integrate-and-fire neuron without external stimuli, enhancing energy efficiency and simplifying fabrication for spintronic neural networks.

Contribution

It demonstrates a novel shape-based design of magnetic tunnel junction neurons that inherently perform leakage, eliminating the need for external stimuli and reducing hardware complexity.

Findings

Shape-based DW drift enables leakage without external stimuli

Design reduces energy consumption and fabrication complexity

Potential for scalable spintronic neural network arrays

Abstract

Spintronic devices based on domain wall (DW) motion through ferromagnetic nanowire tracks have received great interest as components of neuromorphic information processing systems. Previous proposals for spintronic artificial neurons required external stimuli to perform the leaking functionality, one of the three fundamental functions of a leaky integrate-and-fire (LIF) neuron. The use of this external magnetic field or electrical current stimulus results in either a decrease in energy efficiency or an increase in fabrication complexity. In this work, we modify the shape of previously demonstrated three-terminal magnetic tunnel junction neurons to perform the leaking operation without any external stimuli. The trapezoidal structure causes shape-based DW drift, thus intrinsically providing the leaking functionality with no hardware cost. This LIF neuron therefore promises to advance the development of spintronic neural network crossbar arrays.