Researchers at NIST have developed a new approach to automatically tune quantum dot devices using machine learning. The method, called “ray-based learning,” uses one-dimensional traces measured in multiple directions to train an algorithm to recognize the relative position of features characterizing each state. This approach can significantly reduce the number of measured points required for tuning, potentially enabling more efficient development of scalable quantum computers.
Justyna Zwolak, a scientist at NIST, has been working on this project. She uses machine learning algorithms and artificial intelligence to enhance and control quantum systems and computing platforms. The goal is to automatically identify stable configurations of electron spins in semiconductor-based quantum computing, eliminating the need for manual tuning and enabling scaling up of quantum computing into larger arrays of quantum dots.
Source: https://www.nist.gov/itl/itl-speakers-bureau-justyna-zwolak
Keywords: Machine Learning, Quantum Dots, Auto-Tuning
