![Institutional Repository [SANDBOX]](images/dias.logo.png "Institutional Repository [SANDBOX]"){kind=logo}
Institutional Repository [SANDBOX]
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/86D3B384-EB1C-4FD6-B956-A562E3A64E11 | ||
| Year | 2020 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Themistoklis-Ioannis Skordias, "Quantum machine learning, applications and implementation in quantum Hardware", Diploma Work, School of Electrical & Computer Engineering,Technical Univesity of Crete, Chania, Greece, 2020 https://doi.org/10.26233/heallink.tuc.86653 | ||
| Appears in Collections |
This thesis deals with the interface of classical and quantum machine learning.We start by introducing the basic principles of quantum computation, the notion of aqubit, single and two qubit gates, entanglement, as well as the workings of some of the basic quantum algorithms such as the Deutsch algorithm. As a next step we discuss in detail the mathematics of the two building blocks of advanced quantum algorithms, the quantum phase estimation and quantum fourier transform. We then proceed by reviewing the classical machine learning methods and more specifically the Principal Component Analysis algorithm used in the reducing the number of features in complex data analytics problems.In the main part of the thesis, we analyze the quantum Principal Component Analysis algorithm, present in detail the required quantum gates, steps, and circuits involved and also discuss the expected speed ups compared to the classical case. In this main part, we also present implementation and comparison of both algorithms using online prototype available quantum computers by IBM Q using the QSkit quantum programming language, as well our own simulators in Python. An example using real data is used to compare the performance in each case, and to illustrate the differences and advantages of the quantum case compared to the classical one.
Copyright © DIAS 2013
