Machine learning enables completely automatic tuning of a quantum device faster than human experts
Device variability is a bottleneck for the scalability of semiconductor quantum devices. Increasing device control comes at the cost of a large parameter space that has to be explored in order to find the optimal operating conditions. We demonstrate a statistical tuning algorithm that navigates this...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Springer Nature
2020
|
| _version_ | 1797090964142030848 |
|---|---|
| author | Moon, H Lennon, DT Kirkpatrick, J van Esbroeck, NMV Camenzind, LC Yu, L Vigneau, F Zumbühl, DM Briggs, GAD Osborne, MA Sejdinovic, D Laird, EA Ares, N |
| author_facet | Moon, H Lennon, DT Kirkpatrick, J van Esbroeck, NMV Camenzind, LC Yu, L Vigneau, F Zumbühl, DM Briggs, GAD Osborne, MA Sejdinovic, D Laird, EA Ares, N |
| author_sort | Moon, H |
| collection | OXFORD |
| description | Device variability is a bottleneck for the scalability of semiconductor
quantum devices. Increasing device control comes at the cost of a large
parameter space that has to be explored in order to find the optimal operating
conditions. We demonstrate a statistical tuning algorithm that navigates this
entire parameter space, using just a few modelling assumptions, in the search
for specific electron transport features. We focused on gate-defined quantum
dot devices, demonstrating fully automated tuning of two different devices to
double quantum dot regimes in an up to eight-dimensional gate voltage space. We
considered a parameter space defined by the maximum range of each gate voltage
in these devices, demonstrating expected tuning in under 70 minutes. This
performance exceeded a human benchmark, although we recognise that there is
room for improvement in the performance of both humans and machines. Our
approach is approximately 180 times faster than a pure random search of the
parameter space, and it is readily applicable to different material systems and
device architectures. With an efficient navigation of the gate voltage space we
are able to give a quantitative measurement of device variability, from one
device to another and after a thermal cycle of a device. This is a key
demonstration of the use of machine learning techniques to explore and optimise
the parameter space of quantum devices and overcome the challenge of device
variability. |
| first_indexed | 2024-03-07T03:26:13Z |
| format | Journal article |
| id | oxford-uuid:b9206697-df0e-43b8-acdf-eb407925c6ca |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T03:26:13Z |
| publishDate | 2020 |
| publisher | Springer Nature |
| record_format | dspace |
| spelling | oxford-uuid:b9206697-df0e-43b8-acdf-eb407925c6ca2022-03-27T05:00:55ZMachine learning enables completely automatic tuning of a quantum device faster than human expertsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:b9206697-df0e-43b8-acdf-eb407925c6caEnglishSymplectic ElementsSpringer Nature 2020Moon, HLennon, DTKirkpatrick, Jvan Esbroeck, NMVCamenzind, LCYu, LVigneau, FZumbühl, DMBriggs, GADOsborne, MASejdinovic, DLaird, EAAres, NDevice variability is a bottleneck for the scalability of semiconductor quantum devices. Increasing device control comes at the cost of a large parameter space that has to be explored in order to find the optimal operating conditions. We demonstrate a statistical tuning algorithm that navigates this entire parameter space, using just a few modelling assumptions, in the search for specific electron transport features. We focused on gate-defined quantum dot devices, demonstrating fully automated tuning of two different devices to double quantum dot regimes in an up to eight-dimensional gate voltage space. We considered a parameter space defined by the maximum range of each gate voltage in these devices, demonstrating expected tuning in under 70 minutes. This performance exceeded a human benchmark, although we recognise that there is room for improvement in the performance of both humans and machines. Our approach is approximately 180 times faster than a pure random search of the parameter space, and it is readily applicable to different material systems and device architectures. With an efficient navigation of the gate voltage space we are able to give a quantitative measurement of device variability, from one device to another and after a thermal cycle of a device. This is a key demonstration of the use of machine learning techniques to explore and optimise the parameter space of quantum devices and overcome the challenge of device variability. |
| spellingShingle | Moon, H Lennon, DT Kirkpatrick, J van Esbroeck, NMV Camenzind, LC Yu, L Vigneau, F Zumbühl, DM Briggs, GAD Osborne, MA Sejdinovic, D Laird, EA Ares, N Machine learning enables completely automatic tuning of a quantum device faster than human experts |
| title | Machine learning enables completely automatic tuning of a quantum device faster than human experts |
| title_full | Machine learning enables completely automatic tuning of a quantum device faster than human experts |
| title_fullStr | Machine learning enables completely automatic tuning of a quantum device faster than human experts |
| title_full_unstemmed | Machine learning enables completely automatic tuning of a quantum device faster than human experts |
| title_short | Machine learning enables completely automatic tuning of a quantum device faster than human experts |
| title_sort | machine learning enables completely automatic tuning of a quantum device faster than human experts |
| work_keys_str_mv | AT moonh machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT lennondt machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT kirkpatrickj machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT vanesbroecknmv machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT camenzindlc machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT yul machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT vigneauf machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT zumbuhldm machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT briggsgad machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT osbornema machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT sejdinovicd machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT lairdea machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts AT aresn machinelearningenablescompletelyautomatictuningofaquantumdevicefasterthanhumanexperts |