Optimizing the quantum stack: a machine learning approach
Doctoral thesis, 2024
This compilation thesis explores the intersection of machine learning and quantum computing, focusing on optimizing quantum systems and exploring use-cases for quantum computers. Motivated by the potential impact of quantum computers, we investigate several key areas.
First, we delve into machine learning and optimization techniques, establishing the foundation for our research. We then explore reinforcement learning, enabling machines to learn through interaction.
Building on these concepts, we investigate variational quantum algorithms as a promising framework for near-term quantum computing. We analyze the quantum approximate optimization algorithm and introduce gradient-based optimization techniques for VQAs, aiming to assess the potential of quantum computing in solving real-world challenges.
Next, we focus on quantum circuit optimization, proposing methods to combine machine learning techniques with the qubit allocation and routing problem. This work aims to narrow the gap between theoretical quantum algorithms and their practical implementation on quantum hardware.
Finally, we focus on quantum error correction, developing a reinforcement learning approach to efficiently decode error syndromes. This demonstrates the potential of machine learning in enhancing the reliability of quantum computations.
Throughout the thesis, we highlight the benefits of using classical machine learning methods to optimize processes on a quantum computer, contributing to the advancement of quantum computing technologies and their practical applications.
First, we delve into machine learning and optimization techniques, establishing the foundation for our research. We then explore reinforcement learning, enabling machines to learn through interaction.
Building on these concepts, we investigate variational quantum algorithms as a promising framework for near-term quantum computing. We analyze the quantum approximate optimization algorithm and introduce gradient-based optimization techniques for VQAs, aiming to assess the potential of quantum computing in solving real-world challenges.
Next, we focus on quantum circuit optimization, proposing methods to combine machine learning techniques with the qubit allocation and routing problem. This work aims to narrow the gap between theoretical quantum algorithms and their practical implementation on quantum hardware.
Finally, we focus on quantum error correction, developing a reinforcement learning approach to efficiently decode error syndromes. This demonstrates the potential of machine learning in enhancing the reliability of quantum computations.
Throughout the thesis, we highlight the benefits of using classical machine learning methods to optimize processes on a quantum computer, contributing to the advancement of quantum computing technologies and their practical applications.
quantum machine learning
quantum information
quantum circuit optimization
Quantum computing
machine learning
variational quantum algorithm
reinforcement learning
optimization
combinatorial optimization
generative neural networks
Kollectorn
Opponent: Dr. Mario Kren, Max Planck Institute for the Science of Light, Erlangen, Germany
Author
David Fitzek
Chalmers, Microtechnology and Nanoscience (MC2), Applied Quantum Physics
Applying quantum approximate optimization to the heterogeneous vehicle routing problem
Scientific Reports,;Vol. 14(2024)
Journal article
David Fitzek, Yi Hong Teoh, Hin Pok Fung, Gebremedhin A. Dagnew, Ejaaz Merali, M. Schuyler Moss, Benjamin MacLellan, and Roger G. Melko, "RydbergGPT"
We stand at the dawn of the third quantum revolution, where quantum computers promise to transform the landscape of computation. This thesis explores the frontier where quantum computing meets machine learning, addressing key challenges in this emerging field.
Quantum computers harness the bizarre properties of quantum physics, offering exponential computational growth with each added qubit. A mere 50-qubit quantum computer can already challenge classical supercomputers. However, realizing this potential is like tuning a grand piano on a ship during a storm — qubits are extremely sensitive to environmental disturbances, easily losing their quantum properties.
While the ultimate goal is large-scale, error-corrected quantum computers with millions of qubits, the path there is fraught with obstacles. This research investigates how machine learning can enhance quantum systems, from optimizing circuits to improving error correction strategies, thereby circumventing some of these obstacles.
By leveraging machine learning techniques, we explore ways to make quantum devices more robust and capable. This work demonstrates novel approaches for using machine learning to design, control, and operate quantum systems more effectively, paving the way for practical quantum computation.
Quantum computers harness the bizarre properties of quantum physics, offering exponential computational growth with each added qubit. A mere 50-qubit quantum computer can already challenge classical supercomputers. However, realizing this potential is like tuning a grand piano on a ship during a storm — qubits are extremely sensitive to environmental disturbances, easily losing their quantum properties.
While the ultimate goal is large-scale, error-corrected quantum computers with millions of qubits, the path there is fraught with obstacles. This research investigates how machine learning can enhance quantum systems, from optimizing circuits to improving error correction strategies, thereby circumventing some of these obstacles.
By leveraging machine learning techniques, we explore ways to make quantum devices more robust and capable. This work demonstrates novel approaches for using machine learning to design, control, and operate quantum systems more effectively, paving the way for practical quantum computation.
Wallenberg Centre for Quantum Technology (WACQT)
Knut and Alice Wallenberg Foundation (KAW 2017.0449, KAW2021.0009, KAW2022.0006), 2018-01-01 -- 2030-03-31.
Areas of Advance
Nanoscience and Nanotechnology
Subject Categories
Computer and Information Science
Physical Sciences
Nano Technology
ISBN
978-91-8103-108-9
Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 5566
Publisher
Chalmers
Kollectorn
Opponent: Dr. Mario Kren, Max Planck Institute for the Science of Light, Erlangen, Germany