A Nanoscale Room-Temperature Multilayer Skyrmionic Synapse for Deep Spiking Neural Networks

Abstract

Magnetic skyrmions have attracted considerable interest, especially after their recent experimental demonstration at room temperature in multilayers. The robustness, nanoscale size and non-volatility of skyrmions have triggered a substantial amount of research on skyrmion-based low-power, ultra-dense nanocomputing and neuromorphic systems such as artificial synapses. Room-temperature operation is required to integrate skyrmionic synapses in practical future devices. Here, we numerically propose a nanoscale skyrmionic synapse composed of magnetic multilayers that enables room-temperature device operation tailored for optimal synaptic resolution. We demonstrate that when embedding such multilayer skyrmionic synapses in a simple spiking neural network (SNN) with unsupervised learning via the spike-timing-dependent plasticity rule, we can achieve only a 78% classification accuracy in the MNIST handwritten data set under realistic conditions. We propose that this performance can be significantly improved to about 98.61% by using a deep SNN with supervised learning. Our results illustrate that the proposed skyrmionic synapse can be a potential candidate for future energy-efficient neuromorphic edge computing.

14 pages, 10 figures

References in corpus (28)

• Spintronics: Fundamentals and applications
• The design and verification of Mumax3
• Observation of room temperature magnetic skyrmions and their current-driven dynamics in ultrathin Co films
• Deep Learning in Spiking Neural Networks
• Spin Transfer Torques in MnSi at Ultra-low Current Densities
• Skyrmions at room temperature : From magnetic thin films to magnetic multilayers
• Additive interfacial chiral interaction in multilayers for stabilization of small individual skyrmion at room temperature
• Room temperature chiral magnetic skyrmion in ultrathin magnetic nanostructures
• Emergent electrodynamics of skyrmions in a chiral magnet
• Neuromorphic computing with multi-memristive synapses
• Magnetic skyrmion artificial synapse for neuromorphic computing
• Neuromorphic electronic circuits for building autonomous cognitive systems
• Spin torque study of the spin Hall conductivity and spin diffusion length in platinum thin films with varying resistivity
• Skyrmion-electronics: Writing, deleting, reading and processing magnetic skyrmions toward spintronic applications
• Room-temperature current-induced generation and motion of sub-100nm skyrmions
• Laser-Induced Magnetic Nanostructures with Tunable Topological Properties
• A magnetic skyrmion as a non-linear resistive element - a potential building block for reservoir computing
• Magnetic skyrmion-based synaptic devices
• Hybrid chiral domain walls and skyrmions in magnetic multilayers
• Path to collapse for an isolated Néel skyrmion
• Ultrafast generation of skyrmionic defects with vortex beams: printing laser profiles on magnets
• Current-Driven Skyrmion Dynamics and Drive-Dependent Skyrmion Hall Effect in an Ultrathin Film
• Chiral Magnetic Skyrmions with Arbitrary Topological Charge ("skyrmionic sacks")
• Hybrid Spintronic-CMOS Spiking Neural Network With On-Chip Learning: Devices, Circuits and Systems
• A neuromorphic systems approach to in-memory computing with non-ideal memristive devices: From mitigation to exploitation
• Skyrmions in magnetic tunnel junctions
• Twisted domain walls and skyrmions in perpendicularly magnetized multilayers
• Technology Aware Training in Memristive Neuromorphic Systems based on non-ideal Synaptic Crossbars

Cited by in corpus (8)

• Quantum materials for energy-efficient neuromorphic computing
• Bilayer-skyrmion-based design of neuron and synapse for spiking neural network
• Neuromorphic weighted sums with magnetic skyrmions
• Voltage-Controlled Skyrmionic Interconnect with Multiple Magnetic Information Carriers
• Materials for Quantum Technologies: a Roadmap for Spin and Topology
• Tailoring energy barriers of Bloch-point-mediated transitions between topological spin textures
• First-principles Prediction of Potential Candidate Materials MCu${}_3$X${}_4$ (M = V, Nb, Ta; X = S, Se, Te) for Neuromorphic Computing
• Positional stability of skyrmions in a racetrack memory with notched geometry