SAFE Quantum Machine Learning with Variational Quantum Classifiers
For safety-critical applications requiring reliable AI, this work shows that variational quantum classifiers can offer a more balanced profile of accuracy, robustness, and explainability than classical models.
The paper proposes a variational quantum classifier with amplitude encoding and a learnable classical pre-encoding layer, achieving competitive predictive performance compared to classical baselines while demonstrating improved robustness to noise and stability under feature removal, as measured by SAFE-AI metrics.
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.