Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:1806.04318 · Physical Review Applied

Compressed Optimization of Device Architectures for Semiconductor Quantum Devices

Adam Frees, John King Gamble, Daniel R. Ward, Robin Blume-Kohout, M.A. Eriksson, Mark G. Friesen, S. N. Coppersmith

Published Feb 25, 2019• 7 citations
arXivHTMLPDF

Abstract

Recent advances in nanotechnology allow very accurate manipulation of small collections of quantum dots (QDs), an important step in semiconductor-based quantum computation. As the number of QDs in a device grows, it becomes increasingly useful to control the system with as few voltage changes as possible. The authors offer a protocol, ``CODA'', both to efficiently identify sparse sets of voltage changes for controlling a quantum system, and to introduce a metric for comparing device designs. As a method for efficiently controlling $a\phantom{\rule{0}{0ex}}n\phantom{\rule{0}{0ex}}d$ designing a device, CODA will impact engineering solutions to the extension of semiconductor-based quantum computation.

DOI
10.1103/physrevapplied.11.024063

Related papers

Compressed optimization of device architectures

Adam Frees, John King Gamble, Daniel R. Ward +4 more

Note: This preprint has been superseded by arXiv:1806.04318. Recent advances in nanotechnology have enabled researchers to control individual quantum mechanical objects with unprecedented accuracy, opening the door for both quantum and extreme-scale conventional computing applications. As these devices become larger and more complex, the ability to design them such that they can be simply controlled becomes a daunting and computationally infeasible task. Here, motivated by ideas from compressed sensing, we introduce a protocol for the Compressed Optimization of Device Architectures (CODA). It leads naturally to a metric for benchmarking device performance and optimizing device designs, and provides a scheme for automating the control of gate operations and reducing their complexity. Because CODA is computationally efficient, it is readily extensible to large systems. We demonstrate the CODA benchmarking and optimization protocols through simulations of up to eight quantum dots in devices that are currently being developed experimentally for quantum computation.

arXiv:1409.3846•Sep 12, 2014•2 citations

Colloquium : Advances in automation of quantum dot devices control

Justyna P. Zwolak, Jacob M. Taylor

Arrays of quantum dots (QDs) are a promising candidate system to realize scalable, coupled qubit systems and serve as a fundamental building block for quantum computers. In such semiconductor quantum systems, devices now have tens of individual electrostatic and dynamical voltages that must be carefully set to localize the system into the single-electron regime and to realize good qubit operational performance. The mapping of requisite QD locations and charges to gate voltages presents a challenging classical control problem. With an increasing number of QD qubits, the relevant parameter space grows sufficiently to make heuristic control unfeasible. In recent years, there has been considerable effort to automate device control that combines script-based algorithms with machine learning (ML) techniques. In this Colloquium, a comprehensive overview of the recent progress in the automation of QD device control is presented, with a particular emphasis on silicon- and GaAs-based QDs formed in two-dimensional electron gases. Combining physics-based modeling with modern numerical optimization and ML has proven effective in yielding efficient, scalable control. Further integration of theoretical, computational, and experimental efforts with computer science and ML holds vast potential in advancing semiconductor and other platforms for quantum computing.

arXiv:2112.09362•Reviews of Modern Physics•Feb 17, 2023•71 citations

Toward optimizing multi-qubit device design and operation.

Adam Frees

MOTIVATIONS AND PROCEDURE• Designing devices which can be simply controlled may become infeasible as device complexity grows1.• A"simple" operation involves a small number of electrodes located near to the quantity of interest.• Using ideas from compressed sensing2 ' 3, we can find the"simplest" operation associated with a desired outcome.The CODA procedure works as follows: 15 nm 25 nm 10 nm 500 nm Metal control gates Operational target

•Apr 10, 2019•0 citations

Advances in Modeling of Noisy Quantum Computers: Spin Qubits in Semiconductor Quantum Dots

Davide Costa, Mario Simoni, Gianluca Piccinini +1 more

The new quantum era is expected to have an unprecedented social impact, enabling the research of tomorrow in several pivotal fields. These perspectives require a physical system able to encode, process and store for a sufficiently long amount of time the quantum information. However, the optimal engineering of currently available quantum computers, which are small and flawed by several non-ideal phenomena, requires an efficacious methodology for exploring the design space. Hence, there is an unmet need for the development of reliable hardware-aware simulation infrastructures able to efficiently emulate the behaviour of quantum hardware that commits to looking for innovative systematic ways, with a bottom-up approach starting from the physical level, moving to the device level and up to the system level. This article discusses the development of a classical simulation infrastructure for semiconductor quantum-dot quantum computation based on compact models, where each device is described in terms of the main physical parameters affecting its performance in a sufficiently easy way from a computational point of view for providing accurate results without involving sophisticated physical simulators, thus reducing the requirements on CPU and memory. The effectiveness of the involved approximations is tested on a benchmark of quantum circuits — in the expected operating ranges of quantum hardware — by comparing the corresponding outcomes with those obtained via numeric integration of the Schrödinger equation. The achieved results give evidence that this work is a step forward towards the definition of a classical simulator of quantum computers.

IEEE Access•Jan 1, 2023•8 citations

Controlled Quantum Operations of a Semiconductor Three-Qubit System

Hai-Ou Li, Gang Cao, Guodong Yu +4 more

In semiconductor-based quantum computing, gate operations beyond the two-qubit limit are important, but have remained extremely challenging. The authors realize gate-voltage control of interqubit couplings in a specially designed three-qubit quantum-dot device. They exercise coherent control of both amplitude and phase of a target qubit using the prepared states of two control qubits, and demonstrate the basic functionalities of the universal Toffoli gate. This initial effort to achieve controlled three-qubit operations provides useful insight for research on multiqubit systems in semiconductor devices.

arXiv:1610.06704•Physical Review Applied•Feb 15, 2018•25 citations

Autotuning of Double-Dot Devices In Situ with Machine Learning

Justyna P. Zwolak, Thomas McJunkin, Sandesh S. Kalantre +7 more

A primary obstacle in scaling up quantum computing platforms based on semiconductor quantum dots (QDs) is the full automation of initialization and control. Using ideas from machine learning (ML), pattern recognition, and optimization, the authors implement an autotuning protocol that needs no human intervention to navigate between QD states in real time. A convolutional neural network identifies QD states from $i\phantom{\rule{0}{0ex}}n$ $s\phantom{\rule{0}{0ex}}i\phantom{\rule{0}{0ex}}t\phantom{\rule{0}{0ex}}u$ measurements; importantly, the network is trained exclusively on simulated data, and the scans used are significantly smaller that in manual tuning, and hence much faster. This development is critical to moving up to larger numbers of quantum dots.

arXiv:1909.08030•Physical Review Applied•Mar 31, 2020•88 citations
Recent advances in nanotechnology allow very accurate manipulation of small collections of quantum dots (QDs), an important step in semiconductor-based quantum computation. As the number of QDs in a device grows, it becomes increasingly useful to control the system with as few voltage changes as possible. The authors offer a protocol, ``CODA'', both to efficiently identify sparse sets of voltage changes for controlling a quantum system, and to introduce a metric for comparing device designs. As a method for efficiently controlling a\phantom{\rule{0}{0ex}}n\phantom{\rule{0}{0ex}}d designing a device, CODA will impact engineering solutions to the extension of semiconductor-based quantum computation.
Note: This preprint has been superseded by arXiv:1806.04318. Recent advances in nanotechnology have enabled researchers to control individual quantum mechanical objects with unprecedented accuracy, opening the door for both quantum and extreme-scale conventional computing applications. As these devices become larger and more complex, the ability to design them such that they can be simply controlled becomes a daunting and computationally infeasible task. Here, motivated by ideas from compressed sensing, we introduce a protocol for the Compressed Optimization of Device Architectures (CODA). It leads naturally to a metric for benchmarking device performance and optimizing device designs, and provides a scheme for automating the control of gate operations and reducing their complexity. Because CODA is computationally efficient, it is readily extensible to large systems. We demonstrate the CODA benchmarking and optimization protocols through simulations of up to eight quantum dots in devices that are currently being developed experimentally for quantum computation.
Arrays of quantum dots (QDs) are a promising candidate system to realize scalable, coupled qubit systems and serve as a fundamental building block for quantum computers. In such semiconductor quantum systems, devices now have tens of individual electrostatic and dynamical voltages that must be carefully set to localize the system into the single-electron regime and to realize good qubit operational performance. The mapping of requisite QD locations and charges to gate voltages presents a challenging classical control problem. With an increasing number of QD qubits, the relevant parameter space grows sufficiently to make heuristic control unfeasible. In recent years, there has been considerable effort to automate device control that combines script-based algorithms with machine learning (ML) techniques. In this Colloquium, a comprehensive overview of the recent progress in the automation of QD device control is presented, with a particular emphasis on silicon- and GaAs-based QDs formed in two-dimensional electron gases. Combining physics-based modeling with modern numerical optimization and ML has proven effective in yielding efficient, scalable control. Further integration of theoretical, computational, and experimental efforts with computer science and ML holds vast potential in advancing semiconductor and other platforms for quantum computing.
The new quantum era is expected to have an unprecedented social impact, enabling the research of tomorrow in several pivotal fields. These perspectives require a physical system able to encode, process and store for a sufficiently long amount of time the quantum information. However, the optimal engineering of currently available quantum computers, which are small and flawed by several non-ideal phenomena, requires an efficacious methodology for exploring the design space. Hence, there is an unmet need for the development of reliable hardware-aware simulation infrastructures able to efficiently emulate the behaviour of quantum hardware that commits to looking for innovative systematic ways, with a bottom-up approach starting from the physical level, moving to the device level and up to the system level. This article discusses the development of a classical simulation infrastructure for semiconductor quantum-dot quantum computation based on compact models, where each device is described in terms of the main physical parameters affecting its performance in a sufficiently easy way from a computational point of view for providing accurate results without involving sophisticated physical simulators, thus reducing the requirements on CPU and memory. The effectiveness of the involved approximations is tested on a benchmark of quantum circuits — in the expected operating ranges of quantum hardware — by comparing the corresponding outcomes with those obtained via numeric integration of the Schrödinger equation. The achieved results give evidence that this work is a step forward towards the definition of a classical simulator of quantum computers.
In semiconductor-based quantum computing, gate operations beyond the two-qubit limit are important, but have remained extremely challenging. The authors realize gate-voltage control of interqubit couplings in a specially designed three-qubit quantum-dot device. They exercise coherent control of both amplitude and phase of a target qubit using the prepared states of two control qubits, and demonstrate the basic functionalities of the universal Toffoli gate. This initial effort to achieve controlled three-qubit operations provides useful insight for research on multiqubit systems in semiconductor devices.
A primary obstacle in scaling up quantum computing platforms based on semiconductor quantum dots (QDs) is the full automation of initialization and control. Using ideas from machine learning (ML), pattern recognition, and optimization, the authors implement an autotuning protocol that needs no human intervention to navigate between QD states in real time. A convolutional neural network identifies QD states from $i\phantom{\rule{0}{0ex}}n$ $s\phantom{\rule{0}{0ex}}i\phantom{\rule{0}{0ex}}t\phantom{\rule{0}{0ex}}u$ measurements; importantly, the network is trained exclusively on simulated data, and the scans used are significantly smaller that in manual tuning, and hence much faster. This development is critical to moving up to larger numbers of quantum dots.