.png){kind=icon}
.png){kind=icon}
Return
Qsun: an open-source platform towards practical quantum machine learning applications
.png){kind=icon}
OA
AI
DOI：10.1088/2632-2153/ac5997.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Currently, quantum hardware is restrained by noises and qubit numbers. Thus, a quantum virtual machine (QVM) that simulates operations of a quantum computer on classical computers is a vital tool for developing and testing quantum algorithms before deploying them on real quantum computers. Various variational quantum algorithms (VQAs) have been proposed and tested on QVMs to surpass the limitations of quantum hardware. Our goal is to exploit further the VQAs towards practical applications of quantum machine learning (QML) using state-of-the-art quantum computers. In this paper, we first introduce a QVM named Qsun, whose operation is underlined by quantum state wavefunctions. The platform provides native tools supporting VQAs. Especially using the parameter-shift rule, we implement quantum differentiable programming essential for gradient-based optimization. We then report two tests representative of QML: quantum linear regression and quantum neural network.
Keywords:
quantum virtual machine
quantum machine learning
quantum differentiable programming
quantum linear regression
quantum neural network
Journal
M
IF:
4.6
Papers:
1.1K
Citations:
3.4K