Bulletin of the American Physical Society
APS March Meeting 2019
Volume 64, Number 2
Monday–Friday, March 4–8, 2019; Boston, Massachusetts
Session X34: Computer-aided Tune-up and Calibration of Semiconductor QubitsInvited
|
Hide Abstracts |
Sponsoring Units: DQI Chair: Matthew Reed, HRL Laboratories, LLC Room: BCEC 205A |
Friday, March 8, 2019 8:00AM - 8:36AM |
X34.00001: True machine learning for quantum dot tune-up Invited Speaker: Justyna Zwolak Over the past decade, machine learning (ML) techniques have revolutionized how scientific research is done, from designing new materials to finding significant events in particle physics to assisting drug discovery. Recently, we added to this list by showing how an ML algorithm, combined with optimization routines and a robust but simple physics simulator, can assist experimental efforts in tuning semiconductor quantum dot devices. In particular, we demonstrated that deep convolutional neural networks (CNNs) can be used to characterize the state of single and double quantum dots based on measurements of a current-gate voltage transport characteristics or via the conductance of a nearby charge sensor [1,2]. Our approach provides a paradigm for fully-automated experimental initialization via a closed-loop system that does not rely on human intuition and experience. In this talk, I will discuss how our approach works in the experimental setting, and consider extensions of the few-dot tuning problem -- a low dimensional problem -- to the many-dot scenario [3,4] where CNNs are likely to fail. Our approach recovers the geometry of higher-dimensional spaces using 1D traces ("rays") to "fingerprint" a given state in order to differentiate between various state configurations. Moreover, it not only allows to automate the recognition of states but also to reduce the number of measurements required for tuning. |
Friday, March 8, 2019 8:36AM - 9:12AM |
X34.00002: Computer-assisted quantum dot tuning for quantum computation and simulation Invited Speaker: Lieven Vandersypen Much progress was made using electrostatically defined semiconductor quantum dots as the basis for quantum computation and simulation. In these experiments, a desired potential landscape is formed in a two-dimensional electron gas by individually adjusting the voltages applied to multiple gate electrodes. As a result, the filling of each dot, its electrochemical potential and its tunnel coupling to neighbouring dots or to reservoirs are precisely controlled. Next, control signals for initialization, manipulation and read-out of individual charges and spins are calibrated. Despite this encouraging progress, the field could move faster if tuning and calibration tasks were more extensively assisted or taken over by computer automation. We started work several years ago on automating the initial tuning of coupled quantum dots to the few-electron regime [1] and the subsequent setting of the interdot tunnel coupling [2]. In parallel, we also invested in efficient (fast) measurement methods and the use of virtual gates [3] that allow real-time manual tuning and speed up computer-assisted tuning as well. Currently, our focus is on automating repetitive tasks such as calibration or compensation for background charge switches. I will present our past and present tuning efforts and summarize recent physics experiments on 2x2 and 8x1 quantum dot arrays that are enabled by these tuning methods. |
Friday, March 8, 2019 9:12AM - 9:48AM |
X34.00003: Automated Tuning of Tunnel Couplings and Gate Operations for Semiconductor Spin Qubits Invited Speaker: Pascal Cerfontaine Quantum computers require large numbers of qubits with access to high-fidelity initialization, control and readout. While spin qubits in gate-defined quantum dots have demonstrated many of these properties, they need to be individually tuned by adjusting the voltages applied to electrostatic gates. This is a manual and time-consuming process [1] which will become unfeasible for large arrays of qubits. To address this challenge we present an iterative machine-learning algorithm [2] for automated fine-tuning of quantum dots. We estimate the system response by Bayesian updates and thus reduce the number of required measurements. Benchmarks show that the algorithm is able to adjust the tunnel and lead couplings in a singlet-triplet (ST) qubit in GaAs within 5 iterations. |
Friday, March 8, 2019 9:48AM - 10:24AM |
X34.00004: Strategies for Automated Tune-up of Quantum Dot Arrays Invited Speaker: Seán M Meenehan As quantum dot qubit devices grow beyond two or three dots, computer-aided tools for calibration and control become essential. The first stage in tuning up any quantum dot spin qubit is determining an appropriate set of voltage biases such that all dots are loaded in the desired charge configuration, a task which is already complex at the three-dot level. In this talk we describe an automated process, based on single-dot measurements, simple image analysis, and linear compensation modeling of the device, which enables estimation of arbitrary charge states for a multi-dot system. Crucially, the desired bias state is estimated in a way that keeps all inter-dot tunneling rates within an acceptable range. This ensures relevant charge transitions remain visible, even for dots far away from the electron reservoir, and allows navigation of subsequent charge stability diagrams using image processing techniques. We demonstrate that this method can fully automate tuning of a six dot Si/SiGe device into a configuration appropriate for exchange-only qubit operation, including location of appropriate bias windows for spin preparation and measurement, and a validating demonstration of coherent spin rotations. |
Friday, March 8, 2019 10:24AM - 11:00AM |
X34.00005: Efficiently measuring and tuning quantum devices using machine learning Invited Speaker: Natalia Ares Fulfilling the promise of quantum technologies requires to be able to measure and tune several devices; fault-tolerant factorization using a surface code will require ~108 physical qubits. A long-term approach, based on the success of integrated circuits, is to use electron spins in semiconducting devices. A major obstacle to creating large circuits in this platform is device variability. It is very time consuming to fully characterize and tune each of these devices and this task will rapidly become intractable for humans without the aid of automation. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700