Neuromorphic computing with nanoscale spintronic oscillators

Spoken-digit recognition using a nanoscale spintronic oscillator that mimics the behaviour of neurons demonstrates the potential of such oscillators for realizing large-scale neural networks in future hardware. Neuromorphic computing takes the exceptional information processing capabilities of the biological brain as inspiration and attempts to build artificial neurons, synapses and networks for tackling specific tasks that are challenging or energy-intensive for regular computers, such as recognizing images and patterns in sensory signals. Julie Grollier and colleagues use magnetic nanoscale oscillators to mimic the nonlinear oscillating behaviour of neurons and test the capability of such devices to recognize audio signals. The system was trained to recognize spoken digits from five different voices from a benchmark database and could do so with accuracy comparable to state-of-the-art machine learning. The work opens a new direction for chip-based, low-power, brain-like information processing. Neurons in the brain behave as nonlinear oscillators, which develop rhythmic activity and interact to process information1. Taking inspiration from this behaviour to realize high-density, low-power neuromorphic computing will require very large numbers of nanoscale nonlinear oscillators. A simple estimation indicates that to fit 108 oscillators organized in a two-dimensional array inside a chip the size of a thumb, the lateral dimension of each oscillator must be smaller than one micrometre. However, nanoscale devices tend to be noisy and to lack the stability that is required to process data in a reliable way. For this reason, despite multiple theoretical proposals2,3,4,5 and several candidates, including memristive6 and superconducting7 oscillators, a proof of concept of neuromorphic computing using nanoscale oscillators has yet to be demonstrated. Here we show experimentally that a nanoscale spintronic oscillator (a magnetic tunnel junction)8,9 can be used to achieve spoken-digit recognition with an accuracy similar to that of state-of-the-art neural networks. We also determine the regime of magnetization dynamics that leads to the greatest performance. These results, combined with the ability of the spintronic oscillators to interact with each other, and their long lifetime and low energy consumption, open up a path to fast, parallel, on-chip computation based on networks of oscillators.