Optimization of quantum-dot qubit fabrication via machine learning

Abstract

Precise nanofabrication represents a critical challenge to developing semiconductor quantum-dot qubits for practical quantum computation. Here, we design and train a convolutional neural network to interpret scanning electron micrographs and quantify qualitative features affecting device functionality. The high-throughput strategy is exemplified by optimizing a model lithographic process within a five-dimensional design space and by demonstrating a robust approach to address lithographic proximity effects. The results emphasize the benefits of machine learning for developing stable processes, shortening development cycles, and enforcing quality control during qubit fabrication.

Toward Robust Autotuning of Noisy Quantum dot Devices

“…Recently, many advances have been made towards automated calibration of QD devices [7][8][9][10]. Automated methods have been used to tackle many stages of the calibration process, from understanding fabrication results [11] and coarse device tune-up [7,9,10,[12][13][14][15], to fine calibrations of device parameters [8,16]. The techniques used for automation follow two main schools of thought: script-based algorithms and machine learning (ML) methods.…”

Machine Learning-Assisted Manipulation and Readout of Molecular Spin Qubits

“…in phase estimation problems [11], in sensor calibration [12] and in the readout of trapped-ions qubits [13]. Semicondutor-based quantum dots have benefited from ML in their fabrication processes [14], in the automatic search and tuning of their working points [15][16][17] and in their measurement [18]. Similar advantages have been recently reported also on the design [19], the quantum optimal control [20] and the readout [21][22][23] of superconducting qubits.…”

Colloquium: Advances in automation of quantum dot devices control

“…This is particularly important for high-throughput production, where individual wafers may contain hundreds or even thousands of devices. A path towards automated lithographic quality assesment of QD devices has recently been put forward by Mei et al (2021). The proposed control system relays on convolutional neural networks (CNN) applied to scanning electron microscope (SEM) micrographs collected in-line to assess the device usability based on detection of certain fabrication defects (e.g., particle contamination, proper exposure).…”