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

Chen, Runze; Li, Chen; Liu, Yu; Miles, J.J.; Indiveri, Giacomo; Furber, Steve; Pavlidis, Vasilis F.; Moutafis, Christoforos

doi:10.1103/physrevapplied.14.014096

Cited by 30 publications

(21 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“… 11 , 51 With regards to edge annihilation of quasiparticles, as often reported in the literature, using high- K materials at the nanotrack edges can also prevent this from happening. 20 , 27 , 28 Moreover, the latter concern has already been considered in section 2.2 , e.g., by controlling the electrical current density below 4 MA cm –2 to prevent the magnetic quasiparticles from annihilating at the edge.…”

Section: Resultsmentioning

confidence: 99%

Voltage-Controlled Skyrmionic Interconnect with Multiple Magnetic Information Carriers

Chen

Liu

2022

ACS Appl. Mater. Interfaces

Self Cite

View full text Add to dashboard Cite

Magnetic skyrmions have been in the spotlight since they were observed in technologically relevant systems at room temperature. More recently, there has been increasing interest in additional quasiparticles that may exist as stable/metastable spin textures in magnets, such as the skyrmionium and the antiskyrmionite (i.e., a skyrmion bag with two skyrmions inside) that have distinct topological characteristics. The next challenge and opportunity, at the same time, is to investigate the use of multiple magnetic quasiparticles as information carriers in a single device for next-generation nanocomputing. In this paper, we propose a spintronic interconnect device where multiple sequences of information signals are encoded and transmitted simultaneously by skyrmions, skyrmioniums, and antiskyrmionites. The proposed spintronic interconnect device can be pipelined via voltage-controlled magnetic anisotropy (VCMA) gated synchronizers that behave as intermediate registers. We demonstrate theoretically that the interconnect throughput and transmission energy can be effectively tuned by the VCMA gate voltage and appropriate electric current pulses. By carefully adjusting the device structure characteristics, our spintronic interconnect device exhibits comparable energy efficiency with copper interconnects in mainstream CMOS technologies. This study provides fresh insight into the possibilities of skyrmionic devices in future spintronic applications.

show abstract

Section: Resultsmentioning

confidence: 99%

Voltage-Controlled Skyrmionic Interconnect with Multiple Magnetic Information Carriers

Chen

Liu

2022

ACS Appl. Mater. Interfaces

Self Cite

View full text Add to dashboard Cite

show abstract

“…113 Memristive behavior is obtained by modifying the magnetization texture to obtain gradual switching via spin-torque 114,115 or spin-orbit torques. 116 Domain walls 117 or skyrmions 118,119 imitate neurostransmitter nucleation and propagation in these structures. Optical excitations 22 and antiferromagnetic dynamics 120 open the path to ultrafast, terahertz, spintronic synaptic devices.…”

Section: B Magnetization Dynamicsmentioning

confidence: 99%

Quantum materials for energy-efficient neuromorphic computing

Hoffmann,

Ramanathan,

Grollier

et al. 2022

Preprint

View full text Add to dashboard Cite

Neuromorphic computing approaches become increasingly important as we address future needs for efficiently processing massive amounts of data. The unique attributes of quantum materials can help address these needs by enabling new energy-efficient device concepts that implement neuromorphic ideas at the hardware level. In particular, strong correlations give rise to highly non-linear responses, such as conductive phase transitions that can be harnessed for short and long-term plasticity. Similarly, magnetization dynamics are strongly non-linear and can be utilized for data classification. This paper discusses select examples of these approaches, and provides a perspective for the current opportunities and challenges for assembling quantum-material-based devices for neuromorphic functionalities into larger emergent complex network systems.

show abstract

“…Furthermore, the artificial synapse device needs to have well-controlled and highly symmetrical synaptic weights for long-term potentiation (LTP) and long-term depression (LTD) [34,35]. Previous works [36][37][38] propose an anisotropy barrier region in the middle of the nanotrack which cause skyrmions to experience a force when they are in the vicinity of the barrier while crossing from presynapse to postsynapse region, or vice versa. Moreover, the anisotropy barrier can be modulated to emulate the synaptic weight.…”

Section: Introductionmentioning

confidence: 99%

Bilayer-Skyrmion based design of Neuron and Synapse for Spiking Neural Network

Cen¹,

Wang²,

Fong³

2022

Preprint

View full text Add to dashboard Cite

Magnetic skyrmion technology is promising for the next-generation spintronics-based memory and neuromorphic computing due to their small size, non-volatility and low depinning current density. However, the Magnus force originating from the skyrmion Hall effect causes the skyrmion to move along a curved trajectory, which may lead to the annihilation of the skyrmion in a nanotrack during current-induced skyrmion motion. Consequently, circuits utilizing skyrmionic motion need to be designed to limit the impact of the skyrmion Hall effect. In this work, we propose a design of an artificial neuron, and a synapse using the bilayer device consisting of two antiferromagnetically exchange coupled ferromagnetic layers, which achieves robustness against the skyrmion Hall effect by nullifying the Magnus force. Using micromagnetic simulations, we show that the bilayer device can work as an artificial neuron and also as a synapse by modifying its uniaxial anisotropy. We also demonstrate that our proposed skyrmionic synapse has an intrinsic property of highly linear and symmetric weight update, which is highly desirable for the synapse operation. A spiking neural network implemented using our proposed synapse and neuron was simulated and showed to achieve 96.23% accuracy in performing classification on the MNIST handwritten digit dataset.

show abstract

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

Cited by 30 publications

References 60 publications

Voltage-Controlled Skyrmionic Interconnect with Multiple Magnetic Information Carriers

Voltage-Controlled Skyrmionic Interconnect with Multiple Magnetic Information Carriers

Quantum materials for energy-efficient neuromorphic computing

Bilayer-Skyrmion based design of Neuron and Synapse for Spiking Neural Network

Contact Info

Product

Resources

About