Zeyu Fan, Jonathan Wei Zhong Lau, Leong-Chuan Kwek
SciPost Phys. Core 8, 093 (2025) ·
published 18 December 2025
|
· pdf
This study quantifies the effects of quantum noise on the performance of a quantum convolutional neural network (QCNN), building on parallels with classical convolutional neural networks (CNNs), where added Gaussian noise can improve training speed, accuracy, and generalizability. While such benefits are established for classical CNNs, the influence of noise on quantum counterparts remains insufficiently characterized. We specifically examine three types of quantum noise: decoherence, Gaussian noise arising from imperfect quantum gates and experimental error, and systematic noise introduced during input state preparation. Our analysis provides a detailed assessment of how these distinct noise sources affect QCNN operation and outlines considerations for mitigating their impact. Though a QCNN is used as an example in this work, the methods used here can be applied to other quantum machine learning models as well.
SciPost Phys. 12, 122 (2022) ·
published 8 April 2022
|
· pdf
Simulating the dynamics of many-body quantum systems is believed to be one of the first fields that quantum computers can show a quantum advantage over classical computers. Noisy intermediate-scale quantum (NISQ) algorithms aim at effectively using the currently available quantum hardware. For quantum simulation, various types of NISQ algorithms have been proposed with individual advantages as well as challenges. In this work, we propose a new algorithm, truncated Taylor quantum simulator (TQS), that shares the advantages of existing algorithms and alleviates some of the shortcomings. Our algorithm does not have any classical-quantum feedback loop and bypasses the barren plateau problem by construction. The classical part in our hybrid quantum-classical algorithm corresponds to a quadratically constrained quadratic program (QCQP) with a single quadratic equality constraint, which admits a semidefinite relaxation. The QCQP based classical optimization was recently introduced as the classical step in quantum assisted eigensolver (QAE), a NISQ algorithm for the Hamiltonian ground state problem. Thus, our work provides a conceptual unification between the NISQ algorithms for the Hamiltonian ground state problem and the Hamiltonian simulation. We recover differential equation-based NISQ algorithms for Hamiltonian simulation such as quantum assisted simulator (QAS) and variational quantum simulator (VQS) as particular cases of our algorithm. We test our algorithm on some toy examples on current cloud quantum computers. We also provide a systematic approach to improve the accuracy of our algorithm.
