Can we build small neuromorphic chips capable of training deep networks with billions of parameters? This challenge requires hardware neurons and synapses with nanometric dimensions, which can be individually tuned, and densely connected. While nanosynaptic devices have been pursued actively in recent years, much less has been done on nanoscale artificial neurons. In this paper, we show that spintronic nano-oscillators are promising to implement analog hardware neurons that can be densely interconnected through electromagnetic signals. We show how spintronic oscillators maps the requirements of artificial neurons. We then show experimentally how an ensemble of four coupled oscillators can learn to classify all twelve American vowels, realizing the most complicated tasks performed by nanoscale neurons.
@article{arxiv.1904.11240,
title = {Microwave neural processing and broadcasting with spintronic nano-oscillators},
author = {P. Talatchian and M. Romera and S. Tsunegi and F. Abreu Araujo and V. Cros and P. Bortolotti and J. Trastoy and K. Yakushiji and A. Fukushima and H. Kubota and S. Yuasa and M. Ernoult and D. Vodenicarevic and T. Hirtzlin and N. Locatelli and D. Querlioz and J. Grollier},
journal= {arXiv preprint arXiv:1904.11240},
year = {2019}
}