Metraipolis › archive › dc_00bd6bca71e6

Spectral Graph Neural Networks with Hermite Polynomials: A Comprehensive Study — arXiv:2609.28979

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.28979
License (as archived)
CC0-1.0
Archived
2026-10-08T05:25:49.407Z
SHA-256 of the archived text
caf6b3538f541f8b41a34b07a47a707bc3a58176c5249f100f29e1db2df50770
Stable URL of this page
https://metraipolis.com/documento/dc_00bd6bca71e6

How to cite

Spectral Graph Neural Networks with Hermite Polynomials: A Comprehensive Study — arXiv:2609.28979 — arXiv (en), https://arxiv.org/abs/2609.28979 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_00bd6bca71e6 (Metraipolis archive dc_00bd6bca71e6, sha256 caf6b3538f541f8b41a34b07a47a707bc3a58176c5249f100f29e1db2df50770)
BibTeX
@misc{dc00bd6bca71e6,
  title = {Spectral Graph Neural Networks with Hermite Polynomials: A Comprehensive Study — arXiv:2609.28979},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.28979},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_00bd6bca71e6; sha256 caf6b3538f541f8b41a34b07a47a707bc3a58176c5249f100f29e1db2df50770},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.28979. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We study spectral graph neural networks built from Hermite polynomials and propose HermNet, a simple model that combines a nodewise predictor with normalized Hermite propagation. Its sparse recurrence requires neither eigendecomposition nor a learned basis. We distinguish the basic model from optional coordinate calibration, response normalization and Gaussian derivative regularization. Hermite and other complete polynomial bases span the same degree-bounded filter space, but their coordinates can produce different optimization behavior under limited training budgets. We analyze this behavior through spectral signal energy, label sampling, changes in learned features and the bias--variance trade-off of regularization. Controlled synthetic experiments identify a regime in which plain HermNet outperforms matched polynomial-basis alternatives, including with a jointly trained nonlinear predictor. Curvature regularization further improves HermNet when the same functional penalty is available to every comparator. Fixed-predictor controls support the advantage under short training budgets, but longer training removes the plain-model lead. Matched real-data comparisons show accuracy deficits, and architectural and numerical studies identify further limits. Together, the analysis and experiments clarify when Hermite propagation is useful and how calibration and regularization affect its performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.28979 — 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 2026-10-08T04:50:39.304Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).