Nanoscale neural network using non-linear spin-wave interference

TL;DR

This paper introduces a nanoscale neural network leveraging non-linear spin-wave interference, where all functions are performed by spin-wave propagation and scattering, enabling compact and low-power neuromorphic computing.

Contribution

It demonstrates a novel spin-wave based neural network design with integrated nonlinear activation and signal routing, using inverse design via a micromagnetic solver within a machine learning framework.

Findings

Spin-wave interference transitions from linear to nonlinear at high intensities.

Nonlinear regime enhances the computational power of the spin-wave neural network.

The network can be inverse-designed to perform specific input-output mappings.

Abstract

We demonstrate the design of a neural network, where all neuromorphic computing functions, including signal routing and nonlinear activation are performed by spin-wave propagation and interference. Weights and interconnections of the network are realized by a magnetic field pattern that is applied on the spin-wave propagating substrate and scatters the spin waves. The interference of the scattered waves creates a mapping between the wave sources and detectors. Training the neural network is equivalent to finding the field pattern that realizes the desired input-output mapping. A custom-built micromagnetic solver, based on the Pytorch machine learning framework, is used to inverse-design the scatterer. We show that the behavior of spin waves transitions from linear to nonlinear interference at high intensities and that its computational power greatly increases in the nonlinear regime. We envision small-scale, compact and low-power neural networks that perform their entire function in the spin-wave domain.