Quantum Neural Networks: A Comparative Analysis and Noise Robustness Evaluation

TL;DR

This paper compares different hybrid quantum neural networks for image classification on NISQ devices, analyzing their performance and robustness against various quantum noise types, with QuanNN showing superior noise resilience.

Contribution

It provides a comprehensive comparison of HQNN algorithms and evaluates their noise robustness, guiding model selection for noisy quantum environments.

Findings

QuanNN shows greater robustness to quantum noise channels.

Performance varies significantly across different HQNN architectures.

Noise resilience depends on the specific quantum noise type.

Abstract

In current noisy intermediate-scale quantum (NISQ) devices, hybrid quantum neural networks (HQNNs) offer a promising solution, combining the strengths of classical machine learning with quantum computing capabilities. However, the performance of these networks can be significantly affected by the quantum noise inherent in NISQ devices. In this paper, we conduct an extensive comparative analysis of various HQNN algorithms, namely Quantum Convolution Neural Network (QCNN), Quanvolutional Neural Network (QuanNN), and Quantum Transfer Learning (QTL), for image classification tasks. We evaluate the performance of each algorithm across quantum circuits with different entangling structures, variations in layer count, and optimal placement in the architecture. Subsequently, we select the highest-performing architectures and assess their robustness against noise influence by introducing quantum gate noise through Phase Flip, Bit Flip, Phase Damping, Amplitude Damping, and the Depolarizing Channel. Our results reveal that the top-performing models exhibit varying resilience to different noise gates. However, in most scenarios, the QuanNN demonstrates greater robustness across various quantum noise channels, consistently outperforming other models. This highlights the importance of tailoring model selection to specific noise environments in NISQ devices.