Classical Verification and Enhancement of Near-Term Quantum Devices

Cheng, Bin

Classical Verification and Enhancement of Near-Term Quantum Devices

Cheng, Bin

Permalink

Publication Type:: Thesis
Issue Date:: 2023

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (2.41 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Cheng, Bin
dc.date.accessioned	2024-04-16T01:02:33Z
dc.date.available	2024-04-16T01:02:33Z
dc.date.issued	2023
dc.identifier.uri	http://hdl.handle.net/10453/177934
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	Near-term quantum devices are arguably able to perform computational tasks beyond classical capabilities. But a definitive claim of such a quantum computational advantage relies on classical verification. On the other hand, on the road of pursuing quantum advantage on practical tasks, classical techniques to boost the performance of near-term quantum devices are also required. In this thesis, we explore the classical verification and classical enhancement of near-term quantum devices. In the first part of the thesis, we present results on the verification protocols based on the instantaneous quantum polynomial-time (IQP) model, which is a promising model for achieving verifiable quantum advantage on near-term quantum devices. We first study the interplay between IQP circuits, stabilizer formalism and coding theory, and give a characterization of the correlation functions from IQP circuits. Based on this, we give a new IQP-based construction, called the stabilizer scheme, which enriches the scope of IQP-based schemes while maintaining their simplicity and verifiability. To analyze the classical security, we introduce the Hidden Structured Code (HSC) problem as a well-defined mathematical challenge that underlies the stabilizer scheme. We explore a class of attack algorithms based on secret extraction and give evidence of the security of the stabilizer scheme, assuming the hardness of the HSC problem. Moreover, we show that the vulnerability observed in the original IQP verification protocol is primarily attributed to inappropriate parameter choices, which can be naturally rectified with proper parameter settings. In the second part of the thesis, we first present a machine learning approach to quantum error mitigation. We propose the concept of neighborhood learning, and explore the choice of the neighbor circuits and the learning models. Based on our observations, we give an adaptive learning strategy to dynamically construct the neighbor circuits, that achieves a better tradeoff between performance and required resources compared to various quantum error mitigation techniques. Finally, we present the experimental results on simulating large linear cluster states (up to 33 qubits) with only 4 superconducting qubits. Our experiment is based on the circuit-cutting technique, and achieves a better fidelity on 12-qubit linear cluster state than simulating the state directly on a 12-qubit quantum computer.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/177934/1/thesis.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2023 Bin Cheng
dc.rights	au.edu.uts.lib/cph
dc.title	Classical Verification and Enhancement of Near-Term Quantum Devices	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

Near-term quantum devices are arguably able to perform computational tasks beyond classical capabilities. But a definitive claim of such a quantum computational advantage relies on classical verification. On the other hand, on the road of pursuing quantum advantage on practical tasks, classical techniques to boost the performance of near-term quantum devices are also required. In this thesis, we explore the classical verification and classical enhancement of near-term quantum devices. In the first part of the thesis, we present results on the verification protocols based on the instantaneous quantum polynomial-time (IQP) model, which is a promising model for achieving verifiable quantum advantage on near-term quantum devices. We first study the interplay between IQP circuits, stabilizer formalism and coding theory, and give a characterization of the correlation functions from IQP circuits. Based on this, we give a new IQP-based construction, called the stabilizer scheme, which enriches the scope of IQP-based schemes while maintaining their simplicity and verifiability. To analyze the classical security, we introduce the Hidden Structured Code (HSC) problem as a well-defined mathematical challenge that underlies the stabilizer scheme. We explore a class of attack algorithms based on secret extraction and give evidence of the security of the stabilizer scheme, assuming the hardness of the HSC problem. Moreover, we show that the vulnerability observed in the original IQP verification protocol is primarily attributed to inappropriate parameter choices, which can be naturally rectified with proper parameter settings. In the second part of the thesis, we first present a machine learning approach to quantum error mitigation. We propose the concept of neighborhood learning, and explore the choice of the neighbor circuits and the learning models. Based on our observations, we give an adaptive learning strategy to dynamically construct the neighbor circuits, that achieves a better tradeoff between performance and required resources compared to various quantum error mitigation techniques. Finally, we present the experimental results on simulating large linear cluster states (up to 33 qubits) with only 4 superconducting qubits. Our experiment is based on the circuit-cutting technique, and achieves a better fidelity on 12-qubit linear cluster state than simulating the state directly on a 12-qubit quantum computer.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/177934