A superconducting nanowire spiking element for neural networks

Abstract

As the limits of traditional von Neumann computing come into view, the brain's ability to communicate vast quantities of information using low-power spikes has become an increasing source of inspiration for alternative architectures. Key to the success of these largescale neural networks is a power-efficient spiking element that is scalable and easily interfaced with traditional control electronics. In this work, we present a spiking element fabricated from superconducting nanowires that has pulse energies on the order of ~10 aJ. We demonstrate that the device reproduces essential characteristics of biological neurons, such as a refractory period and a firing threshold. Through simulations using experimentally measured device parameters, we show how nanowire-based networks may be used for inference in image recognition, and that the probabilistic nature of nanowire switching may be exploited for modeling biological processes and for applications that rely on stochasticity.

arXiv comment5 main figures; 7 supplemental figures

References in corpus (11)

• Training and Operation of an Integrated Neuromorphic Network Based on Metal-Oxide Memristors
• Josephson junction simulation of neurons
• A superconducting-nanowire 3-terminal electronic device
• Fast Spiking of a Mott VO2-Carbon Nanotube Composite Device
• A nanoCryotron comparator can connect single-flux quantum circuits to conventional electronics
• A compact superconducting nanowire memory element operated by nanowire cryotrons
• Nanoscale superconducting memory based on the kinetic inductance of asymmetric nanowire loops
• Superconducting Neuromorphic Computing Using Quantum Phase-Slip Junctions
• Chaos and relaxation oscillations in spin-torque windmill neurons
• Frequency pulling and mixing of relaxation oscillations in superconducting nanowires
• Bridging the gap between nanowires and Josephson junctions: a superconducting device based on controlled fluxon transfer across nanowires

Cited by in corpus (10)

• Cryogenic Neuromorphic Hardware
• Self-learning Machines based on Hamiltonian Echo Backpropagation
• A Nanocryotron Memory and Logic Family
• Deep Neuromorphic Networks with Superconducting Single Flux Quanta
• An On-Chip Trainable Neuron Circuit for SFQ-Based Spiking Neural Networks
• Hybrid Synaptic Structure for Spiking Neural Network Realization
• Geometric Scaling of the Current-Phase Relation of Niobium Nano-Bridge Junctions
• Contemporary implementations of spiking bio-inspired neural networks
• Attojoule superconducting thermal logic and memories
• Time-tagging data acquisition system for testing superconducting electronics based on an RFSoC and custom analog frontend