Nanoscale neural network using non-linear spin-wave interference

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.

References in corpus (3)

• Physical reservoir computing -- An introductory perspective
• Inverse-design magnonic devices
• Accelerate micromagnetic simulations with GPU programming in MATLAB

Cited by in corpus (19)

• Roadmap on Spin-Wave Computing
• Reconfigurable Training and Reservoir Computing in an Artificial Spin-Vortex Ice via Spin-Wave Fingerprinting
• Inverse-design magnonic devices
• Roadmap for Unconventional Computing with Nanotechnology
• Pattern recognition with a magnon-scattering reservoir
• Reversal of nanomagnets by propagating magnons in ferrimagnetic yttrium iron garnet enabling nonvolatile magnon memory
• Forecasting the outcome of spintronic experiments with Neural Ordinary Differential Equations
• Neuromorphic Overparameterisation and Few-Shot Learning in Multilayer Physical Neural Networks
• Wave-based extreme deep learning based on non-linear time-Floquet entanglement
• Fully Resonant Magneto-elastic Spin-wave Excitation by Surface Acoustic Waves under Conservation of Energy and Linear Momentum
• Handedness-filter and Doppler shift of spin waves in ferrimagnetic domain walls
• Spectrum evolution and chirping of laser-induced spin wave packets in thin iron films
• Exchange spin waves in thin films with gradient composition
• Numerical model for 32-bit magnonic ripple carry adder
• Elongated Skyrmion as Spin Torque Nano-Oscillator and Magnonic Waveguide
• Spin wave diffraction model for perpendicularly magnetized films
• Complete identification of spin-wave eigenmodes excited by parametric pumping in YIG microdisks
• Measuring a population of spin waves from the electrical noise of an inductively coupled antenna
• Unidirectional propagation of spin waves excited by femtosecond laser pulses in a planar waveguide