Stochastic single flux quantum neuromorphic computing using magnetically tunable Josephson junctions

TL;DR

This paper introduces magnetic Josephson junctions for SFQ neuromorphic circuits, enabling tunable, reconfigurable, and stochastic neural firing at high speeds with low power, advancing superconducting neuromorphic computing.

Contribution

The work presents a novel magnetic Josephson junction component that adds tunability and reconfigurability to SFQ neuromorphic circuits, enabling stochastic operation near thermal limits.

Findings

Able to operate at over 10^21 neural firings per second

Achieves approximately 1 W power dissipation

Demonstrates tunable synaptic behavior in superconducting circuits

Abstract

Single flux quantum (SFQ) circuits form a natural neuromorphic technology with SFQ pulses and superconducting transmission lines simulating action potentials and axons, respectively. Here we present a new component, magnetic Josephson junctions, that have a tunablility and re-configurability that was lacking from previous SFQ neuromorphic circuits. The nanoscale magnetic structure acts as a tunable synaptic constituent that modifies the junction critical current. These circuits can operate near the thermal limit where stochastic firing of the neurons is an essential component of the technology. This technology has the ability to create complex neural systems with greater than 10^21 neural firings per second with approximately 1 W dissipation.