Abstract
Hyperbolic neural networks introduce geometric operations that require explicit treatment in relevance propagation. Equivalent geometric realizations can produce different feature attributions, even when local relevance is conserved. We study this problem through Geometric Representation Invariance (GRI), a specialization of Implementation Invariance, and zero-curvature consistency, which requires identity relevance propagation when a geometric module approaches the identity. We propose LRP-radial-all for origin-centered radial modules, treating geometric scaling as modulation and assigning relevance entirely to the signal branch. The rule conserves relevance, is invariant to equivalent radial factorizations, and satisfies zero-curvature consistency, yielding GRI for a specified Poincaré-Lorentz logarithmic-map construction. In contrast, a conservative LRP-half baseline can violate both consistency criteria. Experiments on hyperbolic MNIST, sEEG, and CIFAR-10 classifiers assess attribution fidelity, qualitative explanations, and runtime. LRP-radial-all achieves competitive attribution fidelity across datasets with runtime comparable to Gradient$\times$Input and substantially lower than Integrated Gradients. These findings motivate geometry-aware propagation rules that distinguish relevance conservation from consistency across equivalent computations.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Xiong, P., Li, S., Ding, Y., Schnake, T., & Nakajima, S. (2026). Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation. https://omanscience.com/en/articles/explaining-hyperbolic-neural-networks-via-geometry-aware-relevance-propagation
MLA 9
Xiong, Ping, et al. "Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation." https://omanscience.com/en/articles/explaining-hyperbolic-neural-networks-via-geometry-aware-relevance-propagation.
Chicago (author–date)
Xiong, Ping, Shanglin Li, Yi Ding, Thomas Schnake, and Shinichi Nakajima. 2026. "Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation." https://omanscience.com/en/articles/explaining-hyperbolic-neural-networks-via-geometry-aware-relevance-propagation.
Harvard
Xiong, P., Li, S., Ding, Y., Schnake, T. and Nakajima, S. (2026) 'Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation', Available at: https://omanscience.com/en/articles/explaining-hyperbolic-neural-networks-via-geometry-aware-relevance-propagation.
Vancouver
Xiong P, Li S, Ding Y, Schnake T, Nakajima S. Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation. https://omanscience.com/en/articles/explaining-hyperbolic-neural-networks-via-geometry-aware-relevance-propagation
IEEE
P. Xiong, S. Li, Y. Ding, T. Schnake, and S. Nakajima, "Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation," https://omanscience.com/en/articles/explaining-hyperbolic-neural-networks-via-geometry-aware-relevance-propagation.