AI approaches are ubiquitously used nowadays to enhance efficiency and automize complicated scientific and engineering tasks. Quantum information experiment is one of the best suited scientific efforts for the application of machine learning, where traditionally high-fidelity control of fragile quantum coherence necessarily involves demanding protocols of pre-calibration and intricate tuning of control knobs. In this seminar, I would like to talk about two representative examples of machine learning approaches that my lab develops: (1) reinforced learning-based automatic tuning of individual electron numbers in artificially fabricated electrostatic potential array in semiconductor quantum dots, and (2) deep learning aided individual nuclear spin detection in a diamond. Throughout the talk, I will briefly introduce the main methods used for each type of experiment and discuss what AI tools can automize otherwise cumbersome everyday tasks. I will also discuss current limitations and some possible outlooks for further development.
Meeting Link: To request for meeting link, please write to firstname.lastname@example.org.