Metraipolis › archive › dc_bc88b97ff921
Beyond Peak Backlog: Conditional Energy and Temporal Geometry in Capacity-Constrained Delayed Bandit Optimization — arXiv:2608.16216
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.16216
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T19:41:32.336Z
- SHA-256 of the archived text
bee4e3fa72a1a59c07046b0032a8177443ccb5a5ff49906d2e55f252998e1c44- Stable URL of this page
https://metraipolis.com/documento/dc_bc88b97ff921
How to cite
Beyond Peak Backlog: Conditional Energy and Temporal Geometry in Capacity-Constrained Delayed Bandit Optimization — arXiv:2608.16216 — arXiv (en), https://arxiv.org/abs/2608.16216 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bc88b97ff921 (Metraipolis archive dc_bc88b97ff921, sha256 bee4e3fa72a1a59c07046b0032a8177443ccb5a5ff49906d2e55f252998e1c44)
BibTeX
@misc{dcbc88b97ff921,
title = {Beyond Peak Backlog: Conditional Energy and Temporal Geometry in Capacity-Constrained Delayed Bandit Optimization — arXiv:2608.16216},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.16216},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bc88b97ff921; sha256 bee4e3fa72a1a59c07046b0032a8177443ccb5a5ff49906d2e55f252998e1c44},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16216. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
What is the right delay complexity when a learner can track only $C$ pending feedback items and discarded feedback is permanently lost? Existing one-point bandit convex optimization guarantees in this model pay $\sqrt{T\sigma_{\max}}$, where $\sigma_{\max}$ is the peak backlog, although unlimited tracking admits the sharper $\sqrt{d_{\mathrm{tot}}}$ dependence on total delay. We introduce a scheduler-side conditional-energy interface that separates rate adaptation from the one-point perturbation filtration and handles the dependent importance weights created by randomized admission. Under the same semi-clairvoyant oracle and pathwise hard-capacity contract, this yields an untuned learner whose delay term scales as $O(\sqrt{E_C d_{\mathrm{tot}}})$, with only an explicit restart factor $E_C$; a public constant-factor peak bound removes this factor while $d_{\mathrm{tot}}$ remains unknown. Under strong convexity, the same interface yields the temporal cost $H_A(d)=\sum_t \sigma_t/(A+t)$. Two delay vectors with identical delay multisets, $d_{\mathrm{tot}}$, $\sigma_{\max}$, and capacity can nevertheless have polynomially different minimax regret, showing that timing matters under curvature even when aggregate delay summaries agree. Finally, a continuous hard family converts tracking capacity into a zeroth-order query budget and gives a complementary capacity-starvation lower endpoint. The upper bounds require $C\ge \ln T+1$ and do not constitute a complete capacity minimax characterization.
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/2608.16216 — 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 2026-10-18T19:02:44.158Z 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).