Metraipolis › archive › dc_c4e191c576d7

Eigenvalues of the Hessian in Deep Learning: The Origin of Symmetry and Its Breaking — arXiv:2610.09919

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09919
License (as archived)
CC0-1.0
Archived
2026-10-08T12:35:56.853Z
SHA-256 of the archived text
6a18c1a4239849b7ef02fdbeb26eecc68d98c4103906addfd0bb04f4283573e6
Stable URL of this page
https://metraipolis.com/documento/dc_c4e191c576d7

How to cite

Eigenvalues of the Hessian in Deep Learning: The Origin of Symmetry and Its Breaking — arXiv:2610.09919 — arXiv (en), https://arxiv.org/abs/2610.09919 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c4e191c576d7 (Metraipolis archive dc_c4e191c576d7, sha256 6a18c1a4239849b7ef02fdbeb26eecc68d98c4103906addfd0bb04f4283573e6)
BibTeX
@misc{dcc4e191c576d7,
  title = {Eigenvalues of the Hessian in Deep Learning: The Origin of Symmetry and Its Breaking — arXiv:2610.09919},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09919},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c4e191c576d7; sha256 6a18c1a4239849b7ef02fdbeb26eecc68d98c4103906addfd0bb04f4283573e6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09919. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hessian spectra at trained models in deep learning exhibit a persistent pattern: eigenvalues organize into distinct clusters, including a large bulk near zero and a few isolated outliers. This paper shows that a natural account of these spectral phenomena emerges when the original setting is understood as a departure from a nearby, otherwise hidden, highly symmetric reference. Modifications, including changes to the architecture, data distribution, or parameter metric, expose a nearby reference configuration whose Hessian exhibits rich invariances-ones not accounted for by weight symmetries. There, symmetry enables a precise description of the spectra, forcing high-dimensional kernels and eigenvalues of large multiplicity. Returning to the original configuration breaks the Hessian symmetry and thereby produces the observed hierarchy of clusters and outliers. The framework is developed in some generality, with a detailed analysis of three-layer ReLU networks and applications to convolutional, graph, and transformer models, as well as to the NTK. The same mechanism is further shown to yield analogous spectral structures in layerwise Hessians and the Gauss-Newton matrix.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Optimization and Control (math.OC); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.09919 — 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 2026-10-08T12:14:15.288Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).