报告时间：2月23日 周六  下午2:30

报告地点：实验室一楼会议室

报告人：Xin Wang（王欣）City University of Hong Kong（香港城市大学）

报告题目： Application of machine learning methods to quantum control problems

报告摘要： 

     In this talk, I will present results from our recent attempts to apply machine learning methods to quantum control problems. In particular, I will discuss how supervised learning can be used to design composite pulse sequences that are robust against noise [1], while at the same time can be used to measure the noise spectra in a spin-qubit device while used in conjunction with randomized benchmarking [2].
On the other hand, I will demonstrate that reinforcement learning can be used to improve the quantum speed limit of transferring a spin in a chain [3],and can help in efficiently preparing a quantum state [4].
The pros and cons of reinforcement learning compared to other methods commonly used (such as gradient-based ones) are also studied in detail [4].
 
[1] X.-C. Yang, M.-H. Yung, and XW, Phys. Rev. A 97, 042324 (2018).
[2] C. Zhang, and XW, arXiv:1810.07914.
[3] X.-M. Zhang, Z.-W. Cui, XW, and M.-H. Yung, Phys. Rev. A 97, 052333 (2018).
[4] X.-M. Zhang, Z. Wei, R. Asad, X.-C. Yang, and XW, arXiv:1902.02157.

编辑时间：2019-02-19 14:46:42