Support Vector Machines on Noisy Intermediate-Scale Quantum Computers

Abstract: Support vector machine algorithms are considered essential for the implementationof automation in a radio access network. Specifically, they are critical inthe prediction of the quality of user experience for video streaming based ondevice and network-level metrics. Quantum support vector machine (QSVM)is the quantum analogue of the classical support vector machine algorithm,which utilizes the properties of quantum computers to exponentially speed upthe algorithm. This thesis provides an implementation of the QSVMclassificationsystem and its fundamental components, the quantum Fourier transform(QFT) and the Harrow-Hassidim-Lloyd (HHL) algorithms, using the opensourcequantum computing software development kits (SDKs), IBM’s Qiskitand Rigetti’s pyQuil, and real quantum computers that can be accessed by publiccloud service. Moreover, the QSVM classification system is implementedon a superconducting quantum computer, IBMQX2, showing the potential ofthis quantum algorithm to be executed on the Noisy Intermediate-Scale Quantum(NISQ) computers.