Microwave boson sampling
Borja Peropadre Email: Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States Gian Giacomo Guerreschi Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States Joonsuk Huh Mueunjae Institute for Chemistry (MIC), Department of Chemistry, Pohang University of Science and Technology (POSTECH), Pohang 790-784, Korea Alán Aspuru-Guzik Email: Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States
Abstract
The first post-classical computation will most probably be performed not on a universal quantum computer, but rather on a dedicated quantum hardware. A strong candidate for achieving this is represented by the task of sampling from the output distribution of linear quantum optical networks. This problem, known as boson sampling, has recently been shown to be intractable for any classical computer, but it is naturally carried out by running the corresponding experiment. However, only small scale realizations of boson sampling experiments have been demonstrated to date. Their main limitation is related to the non-deterministic state preparation and inefficient measurement step. Here, we propose an alternative setup to implement boson sampling that is based on microwave photons and not on optical photons. The certified scalability of superconducting devices indicates that this direction is promising for a large-scale implementation of boson sampling and allows for more flexible features like arbitrary state preparation and efficient photon-number measurements.
中文速览
想在经典计算机上模拟多个光子通过线性光学网络后的输出分布（即"玻色采样"，boson sampling）在计算上极其困难，而直接做实验却能天然完成这一任务，因此它被视为最有希望率先超越经典计算机的量子装置。然而，现有光学方案受制于光子的非确定性制备和低效探测，难以规模化扩展。这篇文章提出用超导电路中的微波光子来替代光学光子实现玻色采样：利用超导谐振腔存储光子、可调耦合环充当光束分束器、量子非破坏测量完成读出，并借助电路量子电动力学（circuit QED）中的Jaynes-Cummings相互作用确定性地制备任意光子数态。理论分析和已有实验数据表明，上述三个关键步骤——态制备、幺正演化、高效测量——均可在现有超导工艺上高保真度地实现，并具备良好的可扩展性，从而为首个真正可大规模实施的玻色采样实验以及分子振动光谱的量子模拟指明了一条切实可行的路径。
原文 arXiv:1510.08064;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1510.08064v1