Metraipolis › archive › dc_c243f315766b
When a Flatness Proxy Is Not a Function: Robustness Certificates and Training Interventions — arXiv:2609.38540
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38540
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:45:21.819Z
- SHA-256 of the archived text
3b93eefe5f8941f72c1b512ea4d3150372107e52d15129f33b9c861fc20a0420- Stable URL of this page
https://metraipolis.com/documento/dc_c243f315766b
How to cite
When a Flatness Proxy Is Not a Function: Robustness Certificates and Training Interventions — arXiv:2609.38540 — arXiv (en), https://arxiv.org/abs/2609.38540 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c243f315766b (Metraipolis archive dc_c243f315766b, sha256 3b93eefe5f8941f72c1b512ea4d3150372107e52d15129f33b9c861fc20a0420)
BibTeX
@misc{dcc243f315766b,
title = {When a Flatness Proxy Is Not a Function: Robustness Certificates and Training Interventions — arXiv:2609.38540},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38540},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c243f315766b; sha256 3b93eefe5f8941f72c1b512ea4d3150372107e52d15129f33b9c861fc20a0420},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38540. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A valid curvature upper bound need not justify either a robustness certificate or an intervention on an intrinsic predictor property. We demonstrate this distinction for a last-layer relative-flatness proxy used in both settings. First, empirical-risk stationarity does not eliminate pointwise first-order loss terms: at a finite global empirical-risk minimum, the retained certificate expression underestimates a loss increase by over $210\times$. We derive a globally valid, gauge-invariant feature-space repair. Second, common-row softmax shifts preserve predictions and the exact contraction while making the proxy unbounded. Even standard reference-class choices double it on average relative to the centered representation. For a single fixed-feature example with at least three classes, scalar retuning generically cannot align the induced probability updates. Row centering gives the orbit-minimized bound and restores value and full-model gradient invariance under this symmetry. Across 45 paired one-step tests on algorithmic and image models, amplified shifts separate raw-regularized predictors while quotient-regularized predictors remain aligned. Long-horizon CIFAR-10 experiments show substantial, reversible suppression of generalization, while evidence for selective delay after memorization is less consistent. Together, these results show that validity as a curvature upper bound does not by itself justify either inversion into a robustness certificate or differentiation into an intrinsic training intervention. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.38540 — 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 2026-10-07T13:18:10.222Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).