Machine Learning for tuning, controlling, and optimizing semiconductor spin qubits

Quantum computers hold the potential to bring the world into a new quantum age. However, creating and controlling quantum bits (qubits) has turned out to be challenging. In semiconductor spin qubits, the qubit is encoded in the spin degree of freedom of a quantum dot, an electrical potential trap used to confine charge carriers. These quantum dots are controlled with gate voltages which are applied to nanosized gate-electrodes. Despite the vast progress in the quality of the materials hosting these qubits, energizing and tuning semiconductor qubit systems still requires a considerable amount of experience, time, and patience because of device-to-device variations. Here we overcome this by replacing the human operator with automated, AI-based algorithms. Our approach is architecture and material agnostic, which allows us to report results in various spin qubit systems such as Germanium/Silicon core-shell nanowires, silicon finFETs, and Gallium-Arsenide quantum dots.

In the first course tuning step, our machine-learning algorithms find and energize hole and electron quantum dots faster than human experts. Then, supported by a physical model, another algorithm searches a large dimensional parameter space for signatures of spin effects necessary to operate and read out spin qubit systems. Finally, we report on automated quality optimization of an all-electrical hole spin qubit by changing relevant system parameters such as magnetic and electric fields, read-out position, driving strength, and qubit energy.

We believe that such AI-based procedures will be crucial for controlling more extensive and complex spin qubit networks required in a quantum processor.