Read the original document by opening this link in a new tab.
Table of Contents
Abstract
Kurzzusammenfassung
Acknowledgements
Publications
1 Introduction
2 Classical neural networks
3 Quantum information
4 Dissipative quantum neural networks
5 No free lunch theorem
6 Training with graph-structured quantum data
7 Quantum generative adversarial networks
8 Conclusion and outlook
Appendix A Dissipative quantum neural networks
Appendix B Training with graph-structured quantum data
Appendix C Quantum generative adversial networks
Bibliography
Curriculum vitae
Summary
Quantum neural networks is a dissertation that combines quantum computing and machine learning. It introduces dissipative quantum neural networks (DQNNs) designed for quantum learning tasks, capable of universal quantum computation, and with low memory requirements. The work discusses artificial neural networks, quantum information, quantum algorithms, and circuits, as well as the architecture and training algorithm of DQNNs. It demonstrates the generalization behavior of these networks and their implementation on quantum computers. The thesis also explores the quantum no free lunch theorem, extends the applications of DQNNs, and introduces a generative adversarial model. The author expresses gratitude to individuals who contributed to the work and lists publications related to the thesis.