Metraipolis › archive › dc_bcaae18214f3
Exploring Oversmoothing with Householder Matrices — arXiv:2608.12514
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.12514
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T04:27:19.652Z
- SHA-256 of the archived text
0f98e0e63ab53db859d6cd4be5c1467e5f87d668acbb88b12139028b616516e9- Stable URL of this page
https://metraipolis.com/documento/dc_bcaae18214f3
How to cite
Exploring Oversmoothing with Householder Matrices — arXiv:2608.12514 — arXiv (en), https://arxiv.org/abs/2608.12514 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bcaae18214f3 (Metraipolis archive dc_bcaae18214f3, sha256 0f98e0e63ab53db859d6cd4be5c1467e5f87d668acbb88b12139028b616516e9)
BibTeX
@misc{dcbcaae18214f3,
title = {Exploring Oversmoothing with Householder Matrices — arXiv:2608.12514},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.12514},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bcaae18214f3; sha256 0f98e0e63ab53db859d6cd4be5c1467e5f87d668acbb88b12139028b616516e9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.12514. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Deep graph neural networks(GNNs) suffer from oversmoothing- a progressive collapse of node representation towards a low information subspace as network depth increases because the normalized graph propagation operator is repeatedly applied directly to the hidden representations. In this work we study Householder Graph Neural Network (HouseGNN). Rather than updating the hidden state like standard GCN, HouseGNN uses the aggregated neighbourhood message solely to estimate a reflection direction; the node embedding is then updated by a Householder reflector followed by GroupSort, yielding a piecewise orthogonal layer that preserves Euclidean norm at every node and at every depth. We prove three core properties: (i) every internal layer preserves the node-wise Euclidean norm; (ii) the Householder reflector is scale scale and sign-invariant in the message; and (iii) pairwise distance between nodes can change through mismatch between node-wise orthogonal operators. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.12514 — 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 2026-10-19T04:20:35.811Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).