Metraipolis › archive › dc_f64bb79ca777
On the (non-)closedness of the set of ReLU networks — arXiv:2611.09887
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09887
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:41:45.508Z
- SHA-256 of the archived text
39ff303d82bfc23c4fddd1b4d02523710c9fad407090b1601cbc3b3b6235ba59- Stable URL of this page
https://metraipolis.com/documento/dc_f64bb79ca777
How to cite
On the (non-)closedness of the set of ReLU networks — arXiv:2611.09887 — arXiv (en), https://arxiv.org/abs/2611.09887 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f64bb79ca777 (Metraipolis archive dc_f64bb79ca777, sha256 39ff303d82bfc23c4fddd1b4d02523710c9fad407090b1601cbc3b3b6235ba59)
BibTeX
@misc{dcf64bb79ca777,
title = {On the (non-)closedness of the set of ReLU networks — arXiv:2611.09887},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09887},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f64bb79ca777; sha256 39ff303d82bfc23c4fddd1b4d02523710c9fad407090b1601cbc3b3b6235ba59},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09887. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We prove that the set $\mathcal{N}_N$ of ReLU networks with one hidden layer and at most $N\in\mathbb N$ neurons is weakly closed in $W^{1,p}(\Omega)$, for a bounded connected open set $\Omega\subset\mathbb R^d$. Consequently, every coercive, weakly lower semicontinuous energy on this space attains its minimum over $\mathcal{N}_N$. This covers the Deep Ritz method with convex integrands, Sobolev training and Lipschitz learning. Moreover, as the weight decay parameter tends to zero, minimizers of the regularized problem converge, up to subsequences, to minimizers with minimal parameter norm. We also prove that the set of ReLU networks with two hidden layers of bounded widths is not closed in $W^{1,p}(\Omega)$ or in $C(\overline\Omega)$. This disproves a conjecture of Petersen, Raslan and Voigtlaender.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Classical Analysis and ODEs (math.CA); Functional Analysis (math.FA); Numerical Analysis (math.NA); Optimization and Control (math.OC).
Pagina del lavoro: https://arxiv.org/abs/2611.09887 — 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 2026-11-09T22:33:47.166Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).