Quantum computing has the potential to greatly surpass classical computing for specific tasks. Despite the noise challenges in current quantum devices, NISQ (Noisy Intermediate-Scale Quantum) devices are becoming more accessible to research groups and institutions. To successfully deploy quantum computing algorithms on hardware or through simulations, specialized middleware software is crucial for transitioning from theoretical research to practical experiments. This thesis introduces an open-source framework for quantum simulation, hardware control, and characterization: Qibo. Simulation tools are vital for designing and benchmarking quantum algorithms before deploying them on quantum hardware. In this thesis, we demonstrate circuit-based quantum simulation using just-in-time compilation on multiple hardware accelerators, achieving performance on par with state-of-the-art full-state vector simulators. Deploying quantum algorithms on hardware requires a distinct software layer, provided by Qibolab. This library offers a unified framework for controlling the various electronics needed to operate a quantum computer. Additionally, Qibolab handles all necessary tasks to prepare quantum circuits for execution on a fully characterized device, including circuit transpilation and compilation. However, controlling the electronics alone is insufficient for deploying algorithms on quantum hardware, as current devices are noisy and prone to decoherence, leading to parameter drifts. To address these issues, this framework includes Qibocal, a package for calibrating quantum devices, which eases the burden on experimentalists dealing with noisy devices. Finally, to demonstrate the framework’ capabilities, we present applications achieved using Qibo, including a novel approach to enhancing ground state preparation with Variational Quantum Eigensolvers and Double Bracket algorithms.
OPEN-SOURCE MIDDLEWARE FOR QUANTUM COMPUTING / A. Pasquale ; coordinatore. A. Mennella ; supervisore: S. Cazza. Dipartimento di Fisica Aldo Pontremoli, 2024 Dec 04. 37. ciclo, Anno Accademico 2024/2025.
OPEN-SOURCE MIDDLEWARE FOR QUANTUM COMPUTING
A. Pasquale
2024
Abstract
Quantum computing has the potential to greatly surpass classical computing for specific tasks. Despite the noise challenges in current quantum devices, NISQ (Noisy Intermediate-Scale Quantum) devices are becoming more accessible to research groups and institutions. To successfully deploy quantum computing algorithms on hardware or through simulations, specialized middleware software is crucial for transitioning from theoretical research to practical experiments. This thesis introduces an open-source framework for quantum simulation, hardware control, and characterization: Qibo. Simulation tools are vital for designing and benchmarking quantum algorithms before deploying them on quantum hardware. In this thesis, we demonstrate circuit-based quantum simulation using just-in-time compilation on multiple hardware accelerators, achieving performance on par with state-of-the-art full-state vector simulators. Deploying quantum algorithms on hardware requires a distinct software layer, provided by Qibolab. This library offers a unified framework for controlling the various electronics needed to operate a quantum computer. Additionally, Qibolab handles all necessary tasks to prepare quantum circuits for execution on a fully characterized device, including circuit transpilation and compilation. However, controlling the electronics alone is insufficient for deploying algorithms on quantum hardware, as current devices are noisy and prone to decoherence, leading to parameter drifts. To address these issues, this framework includes Qibocal, a package for calibrating quantum devices, which eases the burden on experimentalists dealing with noisy devices. Finally, to demonstrate the framework’ capabilities, we present applications achieved using Qibo, including a novel approach to enhancing ground state preparation with Variational Quantum Eigensolvers and Double Bracket algorithms.File | Dimensione | Formato | |
---|---|---|---|
phd_unimi_R13299.pdf
accesso aperto
Descrizione: tesi completa
Tipologia:
Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione
6.58 MB
Formato
Adobe PDF
|
6.58 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.