Magnetic domain wall based synaptic and activation function generator for neuromorphic accelerators

Abstract

Magnetic domain walls are information tokens in both logic and memory devices, and hold particular interest in applications such as neuromorphic accelerators that combine logic in memory. Here, we show that devices based on the electrical manipulation of magnetic domain walls are capable of implementing linear, as well as programmable nonlinear, functions. Unlike other approaches, domain-wall-based devices are ideal for application to both synaptic weight generators and thresholding in deep neural networks. Prototype micrometer-size devices operate with 8 ns current pulses and the energy consumption required for weight modulation is < 16 pJ. Both speed and energy consumption compare favorably to other synaptic nonvolatile devices, with the expected energy dissipation for scaled 20 nm devices close to that of biological neurons.

arXiv comment24 pages, 5 figures

References in corpus (7)

• Acceleration of Deep Neural Network Training with Resistive Cross-Point Devices
• Interface control of the magnetic chirality in CoFeB|MgO heterosctructures with heavy metal underlayers
• Current-induced domain wall motion in compensated ferrimagnet
• Proposal for an All-Spin Artificial Neural Network: Emulating Neural and Synaptic Functionalities Through Domain Wall Motion in Ferromagnets
• Spin-Based Neuron Model with Domain Wall Magnets as Synapse
• Spin-Orbit Torque Induced Spike-Timing Dependent Plasticity
• STT-SNN: A Spin-Transfer-Torque Based Soft-Limiting Non-Linear Neuron for Low-Power Artificial Neural Networks

Cited by in corpus (15)

• Spin-orbit torque switching of magnetic tunnel junctions for memory application
• Chirality as Generalized Spin-Orbit Interaction in Spintronics
• Domain Wall-Magnetic Tunnel Junction Spin Orbit Torque Devices and Circuits for In-Memory Computing
• Spintronics on Chiral Objects
• Emerging spintronics phenomena and applications
• 3D Interconnected Magnetic Nanowire Networks as Potential Integrated Multistate Memristors
• Tunable stochasticity in an artificial spin network
• Choose your tools carefully: A Comparative Evaluation of Deterministic vs. Stochastic and Binary vs. Analog Neuron models for Implementing Emerging Computing Paradigms
• Orthogonal Spin Current Injected Magnetic Tunnel Junction for Convolutional Neural Networks
• Engineering of Intrinsic Chiral Torques in Magnetic Thin Films Based on the Dzyaloshinskii-Moriya Interaction
• Universal quantum computing based on magnetic domain wall qubits
• Initialization-Free Multistate Memristor: Synergy of Spin-Orbit Torque and Magnetic Fields
• Domain Wall Leaky Integrate-and-Fire Neurons with Shape-Based Configurable Activation Functions
• High-mobility inertial domain walls driven by spin-transfer torque in a ferrimagnetic spinel oxide
• Focused Surface Acoustic Wave induced nano-oscillator based reservoir computing