Neural networks can learn to utilize correlated auxiliary noise

Aida Ahmadzadegan; Petar Simidzija; Ming Li; Achim Kempf

doi:10.1038/s41598-021-00502-4

Neural networks can learn to utilize correlated auxiliary noise

Sci Rep. 2021 Nov 3;11(1):21624. doi: 10.1038/s41598-021-00502-4.

Authors

Aida Ahmadzadegan^{1

2

3}, Petar Simidzija⁴, Ming Li⁵, Achim Kempf^{6

7

8}

Affiliations

¹ Perimeter Institute for Theoretical Physics, Waterloo, ON, N2L 2Y5, Canada. ahmadzadegan.aida@gmail.com.
² ForeQast Technologies Limited, Waterloo, ON, N2L 5M1, Canada. ahmadzadegan.aida@gmail.com.
³ Department of Applied Mathematics, University of Waterloo, Waterloo, ON, N2L 3G1, Canada. ahmadzadegan.aida@gmail.com.
⁴ Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, V6T 1Z4, Canada.
⁵ Cheriton School of Computer Science, University of Waterloo, Waterloo, ON, N2L 3G1, Canada.
⁶ Perimeter Institute for Theoretical Physics, Waterloo, ON, N2L 2Y5, Canada.
⁷ Department of Applied Mathematics, University of Waterloo, Waterloo, ON, N2L 3G1, Canada.
⁸ Institute for Quantum Computing, University of Waterloo, Waterloo, ON, N2L 3G1, Canada.

Abstract

We demonstrate that neural networks that process noisy data can learn to exploit, when available, access to auxiliary noise that is correlated with the noise on the data. In effect, the network learns to use the correlated auxiliary noise as an approximate key to decipher its noisy input data. An example of naturally occurring correlated auxiliary noise is the noise due to decoherence. Our results could, therefore, also be of interest, for example, for machine-learned quantum error correction.

Publication types

Research Support, Non-U.S. Gov't