Metraipolis › archive › dc_39e84cb4f5eb

Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation — arXiv:2609.35128

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35128
License (as archived)
CC0-1.0
Archived
2026-10-07T22:38:52.186Z
SHA-256 of the archived text
fac4ae0c04fc4c8818d6f362b2c31d6c160dc020f39cffe0e54007e21c077830
Stable URL of this page
https://metraipolis.com/documento/dc_39e84cb4f5eb

How to cite

Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation — arXiv:2609.35128 — arXiv (en), https://arxiv.org/abs/2609.35128 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_39e84cb4f5eb (Metraipolis archive dc_39e84cb4f5eb, sha256 fac4ae0c04fc4c8818d6f362b2c31d6c160dc020f39cffe0e54007e21c077830)
BibTeX
@misc{dc39e84cb4f5eb,
  title = {Explaining Hyperbolic Neural Networks via Geometry-Aware Relevance Propagation — arXiv:2609.35128},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35128},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_39e84cb4f5eb; sha256 fac4ae0c04fc4c8818d6f362b2c31d6c160dc020f39cffe0e54007e21c077830},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35128. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.35128 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T22:14:28.366Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).