Metraipolis › archive › dc_d52a24da2750

Branch Geometry and Finite-Radius Sensitivity of Hard-ReLU Training — arXiv:2608.30960

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30960
License (as archived)
CC0-1.0
Archived
2026-10-17T08:39:32.062Z
SHA-256 of the archived text
1fb2b387218dacd04f82cb5e7f500e060c210eccfd021918ff7057f75c56d884
Stable URL of this page
https://metraipolis.com/documento/dc_d52a24da2750

How to cite

Branch Geometry and Finite-Radius Sensitivity of Hard-ReLU Training — arXiv:2608.30960 — arXiv (en), https://arxiv.org/abs/2608.30960 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d52a24da2750 (Metraipolis archive dc_d52a24da2750, sha256 1fb2b387218dacd04f82cb5e7f500e060c210eccfd021918ff7057f75c56d884)
BibTeX
@misc{dcd52a24da2750,
  title = {Branch Geometry and Finite-Radius Sensitivity of Hard-ReLU Training — arXiv:2608.30960},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30960},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d52a24da2750; sha256 1fb2b387218dacd04f82cb5e7f500e060c210eccfd021918ff7057f75c56d884},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30960. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Outer-learning algorithms use infinitesimal sensitivities to propose finite changes to initialization or training parameters. For hard-ReLU training, the derivative of the finite program and the derivative of its flow limit do not by themselves specify the response at a chosen radius. We characterize the intervening regime in which the perturbation radius is proportional to the GD step. Integer event rounding then survives at leading order: smooth Euler bias shifts each discrete phase, and upstream rounding moves downstream branch boundaries. We derive the crossing indices and a uniform endpoint expansion for finitely many separated transverse events in piecewise-$C^2$ dynamics, away from recursive phase boundaries. In contractive affine regions, an explicit remainder and complete branch verification certify finite candidate comparisons. Scalar phase frequencies and a coupled feedback ablation test the mechanism; frozen nonlinear-network experiments show radius-dependent prediction accuracy, including incomplete branch matches and failed-word tails. Together with local AD and uniform flow consistency, the result identifies sufficient response regimes: differentiating training is a choice of perturbation resolution as well as a choice of derivative.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.30960 — 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 2026-10-17T08:13:26.857Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).