Metraipolis › archive › dc_d7d7df4e3611
Operator-Theoretic Generalization Bounds for Multitask Deep Learning — arXiv:2608.15982
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.15982
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T19:41:13.164Z
- SHA-256 of the archived text
8ee140bb8b4a7cdf2f01db5ee7551994d9955cf15b7c8bb93a046c8ca077c124- Stable URL of this page
https://metraipolis.com/documento/dc_d7d7df4e3611
How to cite
Operator-Theoretic Generalization Bounds for Multitask Deep Learning — arXiv:2608.15982 — arXiv (en), https://arxiv.org/abs/2608.15982 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d7d7df4e3611 (Metraipolis archive dc_d7d7df4e3611, sha256 8ee140bb8b4a7cdf2f01db5ee7551994d9955cf15b7c8bb93a046c8ca077c124)
BibTeX
@misc{dcd7d7df4e3611,
title = {Operator-Theoretic Generalization Bounds for Multitask Deep Learning — arXiv:2608.15982},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.15982},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d7d7df4e3611; sha256 8ee140bb8b4a7cdf2f01db5ee7551994d9955cf15b7c8bb93a046c8ca077c124},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15982. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We develop operator-theoretic generalization bounds for deep multi-output function classes by representing network layers as Koopman composition operators on vector-valued reproducing kernel Hilbert spaces. In vector-valued Sobolev RKHSs, we derive Rademacher complexity bounds for invertible and width-expanding injective architectures. The estimates separate the output-coupling contribution, represented by the trace of the task matrix, from the layerwise operator norms, Sobolev symbol ratios, determinant factors, and restriction constants generated by the linear maps. We then analyze a distinct one-dimensional Brownian/Cameron--Martin regime. Using the exact anchored derivative-norm characterization of the vector-valued Brownian RKHS, we obtain layerwise bounds for domain-preserving scalar linear maps and anchored diffeomorphic activations; the corresponding factors scale as $|W_l|^{1/2}$ and $\|\sigma_l'\|_\infty^{1/2}$, respectively, and do not involve Sobolev smoothness exponents. Because the Sobolev and Brownian results concern different hypothesis spaces, neither is asserted to dominate the other uniformly. We additionally formulate shared operator learning across tasks, prove a finite-rank representer theorem, derive the exact finite-dimensional problem for squared loss, and establish a target-transfer bound when the learned operator is obtained independently of the target sample. Synthetic and MNIST studies examine stabilized Sobolev-inspired and Brownian-inspired complexity proxies; these empirical proxies are not evaluations of the proved bounds for rank-deficient architectures.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.15982 — 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 2026-10-18T19:02:44.139Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).