Prof. Natalia Ares

Fully machine learning-driven control and characterisation of quantum devices

Machine learning is rapidly proving indispensable in tuning and characterising quantum devices. By facilitating the exploration of complex high-dimensional parameter spaces, these algorithms not only allow for the identification of optimal operational conditions but also surpass human experts in the characterisation of different operational regimes. I will present the first fully autonomous tuning of a spin qubit. This is a major advancement for the scalability of semiconductor quantum technologies. I will also discuss the robustness of machine learning algorithms across various semiconductor devices, emphasising their role in the comparative analysis of quantum device architectures. I will conclude by demonstrating the potential of machine learning to understand variability in nominally identical devices. By using physics-informed machine learning approaches, we revealed the disorder potential in a quantum dot device, providing insights into device characteristics that were previously inaccessible. I will thus discuss how we can bridge the gap between quantum device simulation and reality.