Metraipolis › archive › dc_0916ca9eb736

Invertible Graph Aggregation Can Increase ReLU Readout Width — arXiv:2611.02274

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02274
License (as archived)
CC0-1.0
Archived
2026-11-03T18:37:15.673Z
SHA-256 of the archived text
f7f5639f7aedb1665ecd1ab34b7932a6331ebd4dc6fd7ba6482593ce35895adf
Stable URL of this page
https://metraipolis.com/documento/dc_0916ca9eb736

How to cite

Invertible Graph Aggregation Can Increase ReLU Readout Width — arXiv:2611.02274 — arXiv (en), https://arxiv.org/abs/2611.02274 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0916ca9eb736 (Metraipolis archive dc_0916ca9eb736, sha256 f7f5639f7aedb1665ecd1ab34b7932a6331ebd4dc6fd7ba6482593ce35895adf)
BibTeX
@misc{dc0916ca9eb736,
  title = {Invertible Graph Aggregation Can Increase ReLU Readout Width — arXiv:2611.02274},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02274},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0916ca9eb736; sha256 f7f5639f7aedb1665ecd1ab34b7932a6331ebd4dc6fd7ba6482593ce35895adf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02274. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

An invertible graph aggregation preserves the complete feature table, but can increase the size of a classifier that reads one node at a time. For every even $N\ge4$, we construct a labeled tree with $2N$ vertices and maximum degree three. A shared one-hidden-layer rectified linear unit (ReLU) classifier needs one hidden unit on the raw scalar features and exactly $N$ units after one lazy neighborhood mean. All aggregate values are distinct, and the mixing matrix has condition number at most 8. We determine the complete error curve for heads with a free affine term. The width separation persists under scalar perturbations. Under a specified bounded uniform-noise model with one nuisance coordinate, every one-unit aggregate head has expected error at least $1/12$, while a fixed $N$-unit head has zero error. Controlled fits compare trained heads with these explicit feasible scores. On Roman-empire and Tolokers, aggregation improves fitted prediction, and retaining raw features helps at the largest tested parameter budget. Training-only diagnostics include conditional output-layer optimality bounds for eight fixed runs.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2611.02274 — 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 2026-11-03T18:12:54.556Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).