Microwave Neural Processing and Broadcasting with Spintronic Nano-Oscillators
(2018) 2018 IEEE International Electron Devices Meeting (IEDM) — Location: San Francisco, CA (1.December.2018)
Files
Talatchian2018_IEDM.pdf
Open Access - Adobe PDF
- 635.92 KB
Details
- Authors
- Abstract
- 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.
- Affiliations
Citations
Talatchian, P., Romera, M., Tsunegi, S., Abreu Araujo, F., Cros, V., Bortolotti, P., Trastoy, J., Yakushiji, K., Fukushima, A., Kubota, H., Yuasa, S., Ernoult, M., Vodenicarevic, D., Hirtzlin, T., Locatelli, N., Querlioz, D., & Grollier, J. (2018). Microwave Neural Processing and Broadcasting with Spintronic Nano-Oscillators. 2018 IEEE International Electron Devices Meeting (IEDM), 36(3), 616-619. https://doi.org/10.1109/iedm.2018.8614585 (Original work published 2018)