Machine-Learning enhanced Quantum State Tomography

In this talk, I shall be covering fundamental details about machine-learning (ML) enhanced quantum state tomography (QST) for squeezed states. Implementation of machine learning architecture with a convolutional neural network will be illustrated and demonstrated through the experimentally measured data generated from squeezed vacuum states [1]. In addition to using the reconstruction model in training a truncated density matrix, we also develop a high-performance, lightweight, and easy-to-install supervised characteristic model by generating the target parameters directly [2]. With the help of machine learning-enhanced quantum state tomography, we also experimentally reconstructed the Wigner’s quantum phase current for the first time [3].  A brief view on quantum ML will also be discussed [4]. At the same time, as a collaborator for LIGO-Virgo-KAGRA gravitational wave network and Einstein Telescope, I will introduce our plan to inject this squeezed vacuum field into the advanced gravitational wave detectors [5]. I will also cover progress in applying such a ML- QST as a crucial diagnostic toolbox for applications with squeezed states, from quantum information process, quantum metrology,  and macroscopic quantum state  generation.