SAFE Quantum Machine Learning with Variational Quantum Classifiers

TL;DR

This paper introduces a variational quantum classifier that leverages amplitude encoding and SAFE-AI metrics, achieving competitive accuracy and enhanced robustness in safety-critical applications.

Contribution

It presents a novel quantum classification model with structured hypothesis class and SAFE-AI evaluation, improving robustness and reliability over classical models.

Findings

Competitive predictive performance compared to classical baselines.

Enhanced robustness to noise and structured feature removal.

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.