SAFE Quantum Machine Learning with Variational Quantum Classifiers

2026-05-15

AI Summary

This work addresses the challenge of simultaneously achieving accuracy, robustness, and interpretability in safety-critical machine learning applications, where conventional models often exhibit uncontrolled sensitivity to input perturbations. The authors propose a novel variational quantum classifier that integrates a learnable classical pre-encoding layer with amplitude encoding, leveraging normalized amplitude embedding and bounded quantum observables to construct a structured and smooth hypothesis class capable of delivering controllable responses to input variations. For the first time, the Cramér–von Mises divergence–based SAFE-AI reliability assessment framework is introduced into quantum machine learning. Experimental results demonstrate that the proposed model matches strong classical baselines in predictive performance while significantly outperforming them in noise robustness and stability under structured feature ablation, yielding a more balanced SAFE reliability profile.

Abstract

We propose a variational quantum classifier operating on high dimensional deep representations via amplitude encoding, stabilized by a learnable classical pre encoding layer.By combining normalized amplitude embeddings with bounded quantum observables, the resulting model induces a structured and smooth hypothesis class with controlled sensitivity to input variations. Model reliability is assessed using SAFE-AI metrics derived from the Cramer von Mises divergence, enabling consistent evaluation across accuracy, robustness, and explainability dimensions. Empirical results show that the proposed quantum model provides competitive predictive performance compared with strong classical baselines while exhibiting a more balanced SAFE reliability profile, with improved robustness to noise and stability under structured feature removal. These findings suggest that variational quantum circuits offer a principled mechanism for stability oriented SAFE learning in safety critical settings.

Problem

• SAFE-AI
• quantum machine learning
• variational quantum classifier
• robustness
• reliability

Innovation

• Variational Quantum Classifier
• Amplitude Encoding
• SAFE-AI Metrics
• Robustness
• Quantum Machine Learning