Extract the Degradation Information in Squeezed States with Machine Learning
透過機器學習萃取量子噪音壓縮態的劣化資訊

Hsien-Yi Hsieh, Yi-Ru Chen, Hsun-Chung Wu, Hua Li Chen, Jingyu Ning, Yao-Chin Huang, Chien-Ming Wu, and Ray-Kuang Lee*
Phys. Rev. Lett. 128, 073604 (2022)
DOI: https://doi.org/10.1103/PhysRevLett.128.073604

In order to leverage the full power of quantum noise squeezing with unavoidable decoherence, a complete understanding of the degradation in the purity of  squeezed light is demanded. By implementing machine learning architecture with a convolutional neural network, we illustrate a fast (less than one second), robust (to the thermal noises), and precise (keeping the fidelity as high as 0.99) quantum state tomography for continuous variables, through the experimentally measured data generated from the balanced homodyne detectors. Our neural network enhanced quantum state tomography provides the metrics to give physical descriptions of every feature observed in the quantum state  with a single-scan measurement and paves a way of exploring large-scale quantum systems in real-time.

即使在不可避免的環境干擾下，完整了解量子噪音壓縮態的純度劣化資訊，將可以讓我們充分發揮量子噪音壓縮的非古典威力。透過卷積神經網路的機器學習實作，我們成功展示既快速，又強健，且精準的量子斷層掃描，和連續變數密度矩陣的重建。此新穎的機器學習強化之量子斷層掃描，不僅只需要單次掃描即可完成量子態資訊萃取，並提供即時探討大規模量子系統的可能性。