Metraipolis › archive › dc_09f5210bc444

Linear Independence of Polynomial Compositions with Bounded Input Degree — arXiv:2610.20825

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20825
License (as archived)
CC0-1.0
Archived
2026-10-16T10:24:58.768Z
SHA-256 of the archived text
f56409f47779860690f5a67b313b179cb7bd979a45e03d89171e4da1df32a377
Stable URL of this page
https://metraipolis.com/documento/dc_09f5210bc444

How to cite

Linear Independence of Polynomial Compositions with Bounded Input Degree — arXiv:2610.20825 — arXiv (en), https://arxiv.org/abs/2610.20825 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09f5210bc444 (Metraipolis archive dc_09f5210bc444, sha256 f56409f47779860690f5a67b313b179cb7bd979a45e03d89171e4da1df32a377)
BibTeX
@misc{dc09f5210bc444,
  title = {Linear Independence of Polynomial Compositions with Bounded Input Degree — arXiv:2610.20825},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20825},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09f5210bc444; sha256 f56409f47779860690f5a67b313b179cb7bd979a45e03d89171e4da1df32a377},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20825. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We prove the weak composition-independence conjectures of Kohn, Marchetti, Massarenti and Mella. The proof uses deformations of rational curves and a single kernel of the linearized equation. Riemann-Roch gives a contradiction from tangent fields with controlled poles. By settling these conjectures, we prove that a neural network with a general polynomial activation function of sufficiently large degree is identifiable.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Algebraic Geometry (math.AG) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.20825 — 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 2026-10-16T08:48:02.841Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).