Angular Gradient Sign Method: Uncovering Vulnerabilities in Hyperbolic Networks
By: Minsoo Jo, Dongyoon Yang, Taesup Kim
Potential Business Impact:
Makes AI smarter by tricking it in new ways.
Adversarial examples in neural networks have been extensively studied in Euclidean geometry, but recent advances in \textit{hyperbolic networks} call for a reevaluation of attack strategies in non-Euclidean geometries. Existing methods such as FGSM and PGD apply perturbations without regard to the underlying hyperbolic structure, potentially leading to inefficient or geometrically inconsistent attacks. In this work, we propose a novel adversarial attack that explicitly leverages the geometric properties of hyperbolic space. Specifically, we compute the gradient of the loss function in the tangent space of hyperbolic space, decompose it into a radial (depth) component and an angular (semantic) component, and apply perturbation derived solely from the angular direction. Our method generates adversarial examples by focusing perturbations in semantically sensitive directions encoded in angular movement within the hyperbolic geometry. Empirical results on image classification, cross-modal retrieval tasks and network architectures demonstrate that our attack achieves higher fooling rates than conventional adversarial attacks, while producing high-impact perturbations with deeper insights into vulnerabilities of hyperbolic embeddings. This work highlights the importance of geometry-aware adversarial strategies in curved representation spaces and provides a principled framework for attacking hierarchical embeddings.
Similar Papers
Geometry-Aware Backdoor Attacks: Leveraging Curvature in Hyperbolic Embeddings
Machine Learning (CS)
Makes AI models easier to trick with secret codes.
Adaptive Hyperbolic Kernels: Modulated Embedding in de Branges-Rovnyak Spaces
Artificial Intelligence
Makes computers better at understanding things with layers.
H4G: Unlocking Faithful Inference for Zero-Shot Graph Learning in Hyperbolic Space
Machine Learning (CS)
Helps computers understand complex connections better.