SAFE machine learning with variational quantum classifiers

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 Cramér–von Mises divergence, enabling consistent evaluation across accuracy, robustness, and explainability dimensions. Empirical results on the single brain tumor MRI classification task show that the proposed quantum model provides competitive predictive performance compared with strong classical baselines while exhibiting a more balanced SAFE reliability profile, with particularly strong robustness when perturbations are applied to the extracted deep feature representation 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.