Metraipolis › archive › dc_f259556d40f2

Spectral Indistinguishability and Learning-Curve Identifiability in Quadratic Networks — arXiv:2610.20678

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20678
License (as archived)
CC0-1.0
Archived
2026-10-16T10:19:56.531Z
SHA-256 of the archived text
d69ae7641966eeaa10983d55d94e8f87d13d02e322314d7abd50917abc6cc9b8
Stable URL of this page
https://metraipolis.com/documento/dc_f259556d40f2

How to cite

Spectral Indistinguishability and Learning-Curve Identifiability in Quadratic Networks — arXiv:2610.20678 — arXiv (en), https://arxiv.org/abs/2610.20678 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f259556d40f2 (Metraipolis archive dc_f259556d40f2, sha256 d69ae7641966eeaa10983d55d94e8f87d13d02e322314d7abd50917abc6cc9b8)
BibTeX
@misc{dcf259556d40f2,
  title = {Spectral Indistinguishability and Learning-Curve Identifiability in Quadratic Networks — arXiv:2610.20678},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20678},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f259556d40f2; sha256 d69ae7641966eeaa10983d55d94e8f87d13d02e322314d7abd50917abc6cc9b8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20678. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We study what spectral summaries and scalar learning curves reveal about centered quadratic networks. We construct analytic positive input densities with exactly Gaussian one- and two-coordinate marginals. Their complete fourth-moment operators are intertwined by an orthogonal map fixing the entire diagonal teacher algebra, so the corresponding anchored spectral measures agree. For selected common teachers with a nonzero cycle invariant, their factorized loss curves differ because the map does not preserve matrix multiplication. Cycle length $r$, which can be arbitrarily large, gives $O(\varepsilon^r)$ contrast uniformly in training time for each fixed construction. For a supplied pair, noisy scalar loss queries require order $\varepsilon^{-2r}$ observations, while bounded initialization interventions with scalar gradient readouts require order $\varepsilon^{-2}$. A shared-input score, equal to a paired-gradient difference, attains the optimal input-sample exponent without knowing the teacher. An explicit family of diverging auxiliary teacher scales instead admits a nonzero finite limiting contrast at coupling $1/\log L$. Conversely, for Gaussian inputs and a rationally independent teacher spectrum, one exact scalar loss curve identifies the initial predictor up to sign conjugation. Within this rationally independent class, with nonzero diagonal inverse-state error in every coordinate and calibrated time and loss scales, it also identifies the unknown teacher spectrum and active dimension, modulo simultaneous rotations. For fixed dimension and distinct unordered teacher sums, dimension-minimal exact observations give generic joint local recovery. Finite exact global bounds hold for a known teacher and, on the active class, under a blind dimension cap. Exact certificates and full-weight replay verify the construction; numerical inverse checks reveal severe ill-conditioning.

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/2610.20678 — 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 2026-10-16T08:48:02.640Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).