Accepted author manuscript, 2.8 MB, PDF document
Available under license: CC BY: Creative Commons Attribution 4.0 International License
Final published version
Licence: CC BY: Creative Commons Attribution 4.0 International License
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - Machine learning enables completely automatic tuning of a quantum device faster than human experts
AU - Moon, H.
AU - Lennon, D. T.
AU - Kirkpatrick, J.
AU - van Esbroeck, N. M.
AU - Camenzind, L. C.
AU - Yu, Liuqi
AU - Vigneau, F.
AU - Zumbühl, Dominik
AU - Briggs, G. Andrew D.
AU - Osborne, M. A.
AU - Sejdinovic, D.
AU - Laird, Edward
AU - Ares, N.
PY - 2020/8/19
Y1 - 2020/8/19
N2 - Variability is a problem for the scalability of semiconductor quantum devices. The parameter space is large, and the operating range is small. Our statistical tuning algorithm searches for specific electron transport features in gate-defined quantum dot devices with a gate voltage space of up to eight dimensions. Starting from the full range of each gate voltage, our machine learning algorithmcan tune each device to optimal performance in a median time of under 70 minutes. This performance surpassed our best human benchmark (although both human and machine performance can be improved). The algorithm is approximately 180 times faster than an automated random search of the parameter space, and is suitable for different material systems and device architectures. Our results yield a quantitative measurement of device variability, from one device to another and after thermal cycling. Our machine learning algorithm can be extended to higher dimensions and other technologies.
AB - Variability is a problem for the scalability of semiconductor quantum devices. The parameter space is large, and the operating range is small. Our statistical tuning algorithm searches for specific electron transport features in gate-defined quantum dot devices with a gate voltage space of up to eight dimensions. Starting from the full range of each gate voltage, our machine learning algorithmcan tune each device to optimal performance in a median time of under 70 minutes. This performance surpassed our best human benchmark (although both human and machine performance can be improved). The algorithm is approximately 180 times faster than an automated random search of the parameter space, and is suitable for different material systems and device architectures. Our results yield a quantitative measurement of device variability, from one device to another and after thermal cycling. Our machine learning algorithm can be extended to higher dimensions and other technologies.
U2 - 10.1038/s41467-020-17835-9
DO - 10.1038/s41467-020-17835-9
M3 - Journal article
VL - 11
JO - Nature Communications
JF - Nature Communications
SN - 2041-1723
M1 - 4161
ER -