Making nature compute for us

Authors

  • Martin Stein Cornell University

DOI:

https://doi.org/10.25250/thescbr.brk667

Keywords:

Neural networks, Physics of computation, Optics, Electronics, Audio

Abstract

Artificial intelligence is all the hype lately. Behind many of the mind-blowing breakthroughs of the past decade is a single workhorse: More compute. As engineers work hard to supply the necessary electronics, researchers are turning to less conventional ideas in hopes of finding the next big thing. We showed how to employ the complex computations nature does, free-of-charge, for neural networks.

 

Author Biography

Martin Stein, Cornell University

PhD candidate

Original article reference

Wright, L. G., Onodera, T., Stein, M. M., Wang, T., Schachter, D. T., Hu, Z., & McMahon, P. L. (2022). Deep physical neural networks trained with backpropagation. Nature, 601(7894), 549–555. https://doi.org/10.1038/s41586-021-04223-6

Downloads

  • URL

Published

2023-01-18

Issue

Vol. 9 No. 1 (2023): TheScienceBreaker

Section

Maths, Physics & Chemistry

License

Some rights reserved 2023 Martin Stein

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.