JGRA: Jacobian Geometry Robustness Assessment in NISQ Noise-Aware Quantum Neural Networks

2026-06-08

AI Summary

This work addresses the challenge of evaluating the robustness of quantum neural networks (QNNs) to parameter perturbations in noisy NISQ devices. To this end, it introduces, for the first time, a Jacobian-based geometric analysis combined with entropy-matched noise calibration and noise-aware training to construct a geometric descriptor that captures the relationship between model architecture and noisy inference behavior. This descriptor enables accurate prediction of QNN performance under unseen noise conditions without requiring retraining, thereby offering a novel and practical tool for assessing the reliability of quantum models in the NISQ era.

Abstract

The NISQ era places stringent constraints on quantum computation, where noise and decoherence fundamentally limit performance. In classical deep learning, model robustness and resilience to perturbations are well studied: deep neural networks (DNNs) maintain high performance despite pruning, noise injection, and structural perturbations due to inherent redundancy in their representations. A central challenge in quantum machine learning is to transfer this notion of robustness to quantum neural networks (QNNs) under realistic NISQ noise. While classical deep learning exhibits robustness through structural redundancy, analogous principles for QNNs remain underdeveloped. We propose JGRA: a framework for assessing robustness in noise-aware QNNs via Jacobian geometry, capturing model sensitivity to parameter perturbations induced by noise. Our method includes entropy-matched noise calibration, noise-aware training, and noise-conditioned Jacobian extraction, yielding geometric descriptors that link clean-regime structure to noisy inference behaviour. We also empirically demonstrate that these descriptors encode predictive information about robustness under unseen noise.

Problem

• quantum neural networks
• NISQ
• robustness
• noise
• Jacobian geometry

Innovation

• Jacobian geometry
• quantum neural networks
• NISQ noise
• robustness assessment
• noise-aware training