PN-QNN: Harnessing Physical Noise as a Native Regularizer in Photonic Hybrid Quantum Neural Networks

2026-07-22

AI Summary

This work proposes harnessing physical noise inherent in photonic quantum hardware not as a detrimental factor but as a native regularization mechanism to enhance the generalization capability of hybrid quantum–classical neural networks. A seven-parameter physical noise model is constructed using Perceval, and its configuration is jointly optimized via a genetic algorithm. The study establishes, for the first time, an equivalence between this noise-induced regularization and Tikhonov regularization. Empirical validation on the Quandela Perceval simulator within the MerLin framework demonstrates accuracy improvements of 0.82% and 1.45% on the Iris and Digits datasets, respectively, while a 1.21% decrease on MNIST highlights the task-dependent nature of the effect. By treating noise as a free regularizer, this approach offers a novel paradigm for effectively leveraging noisy quantum hardware.

Abstract

Physical noise in near-term quantum hardware is usually treated as a nuisance to suppress. We ask whether it can instead act as a hardware-native regularizer for photonic hybrid quantum-classical neural networks (PHQCNNs), analogous to noise-injection regularization in classical deep learning. Using Quandela's Perceval simulator and the MerLin framework, we build PHQCNNs for Iris, Digits, and MNIST and inject Perceval's seven-parameter physical noise model directly into training. A genetic algorithm searches the six continuous noise dimensions and 1 boolean parameter to find, per dataset, the configuration maximizing validation accuracy, compared against a noiseless baseline across five seeds. GA-tuned noise yields modest accuracy gains on Iris (+0.82pp) and Digits (+1.45pp), but a clear degradation on MNIST (-1.21pp). Per-parameter sweeps show that no individual noise parameter is consistently beneficial, motivating the joint search, while a second-order loss expansion shows that physical noise induces a Tikhonov-like regularization term whose effect is dataset-dependent. Physical photonic noise can thus act as a free regularizer, but not universally.

Problem

• physical noise
• quantum neural networks
• regularization
• photonic hardware
• noise injection

Innovation

• physical noise as regularizer
• photonic quantum neural networks
• noise-injection regularization
• genetic algorithm optimization
• Tikhonov-like regularization