Feature‐Level Analysis and Adversarial Transfer in Rotationally Equivariant Quantum Machine Learning

ABSTRACT Group‐equivariant quantum models are designed to exploit symmetry and can improve trainability, but it remains unclear how symmetry constraints shape their adversarial robustness. We study this question through a feature‐level analysis of equivariant quantum models in a classical‐to‐quantum transfer attack setting. Under equivariance with an invariant readout, predictions depend only on the group‐twirled input, which identifies the symmetry‐invariant information accessible to the model together with a complementary uninformative subspace. Specializing this framework to a rotationally equivariant quantum model, we derive an explicit characterization of the accessible information in terms of rotation‐invariant image statistics distributed across distinct symmetry sectors. Using targeted input transformations, we identify which of these statistics are actually relied upon for classification across several datasets. Our numerical results show that equivariance alone does not guarantee transfer robustness: even within the restricted invariant feature space, the model can rely on brittle statistics, particularly ring‐averaged intensities in the rotationally equivariant model, that remain vulnerable to classical transfer attacks. Guided by this analysis, we numerically demonstrate that suppressing the symmetry sector associated with the brittle feature substantially improves transfer robustness. Together, our analytical framework and numerical results provide a systematic approach for investigating how symmetry‐dependent features relate to adversarial transfer robustness in equivariant quantum machine learning models.