Adiabatic Superconducting Artificial Neural Network: Basic Cells

TL;DR

This paper introduces adiabatic superconducting cells functioning as neurons and synapses in a multilayer perceptron, offering a compact, energy-efficient hardware implementation with inherent nonlinearity and flux-based signal processing.

Contribution

It presents a novel superconducting cell design for neural network components that are simple, energy-efficient, and capable of implementing common activation functions.

Findings

Cells operate with just one or two Josephson junctions.

Neurons can perform one-shot sigmoid and hyperbolic tangent calculations.

Synapses support positive and negative transfer coefficients in a specific range.

Abstract

We consider adiabatic superconducting cells operating as an artificial neuron and synapse of a multilayer perceptron (MLP). Their compact circuits contain just one and two Josephson junctions, respectively. While the signal is represented as magnetic flux, the proposed cells are inherently nonlinear and close-to-linear magnetic flux transformers. The neuron is capable of providing a one-shot calculation of sigmoid and hyperbolic tangent activation functions most commonly used in MLP. The synapse features by both positive and negative signal transfer coefficients in the range ~ (-0.5,0.5). We briefly discuss implementation issues and further steps toward multilayer adiabatic superconducting artificial neural network which promises to be a compact and the most energy-efficient implementation of MLP.