Metraipolis › archive › dc_9536c141d6fa
ZoAQ: Adaptive Zeroth-Order Querying via Query-Reuse Coupling — arXiv:2609.22115
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.22115
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T07:05:57.871Z
- SHA-256 of the archived text
5d3f0c1fd3e9ee68442c9a6fca7bf3b3916ead26279ac52d9ea90055c967308e- Stable URL of this page
https://metraipolis.com/documento/dc_9536c141d6fa
How to cite
ZoAQ: Adaptive Zeroth-Order Querying via Query-Reuse Coupling — arXiv:2609.22115 — arXiv (en), https://arxiv.org/abs/2609.22115 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9536c141d6fa (Metraipolis archive dc_9536c141d6fa, sha256 5d3f0c1fd3e9ee68442c9a6fca7bf3b3916ead26279ac52d9ea90055c967308e)
BibTeX
@misc{dc9536c141d6fa,
title = {ZoAQ: Adaptive Zeroth-Order Querying via Query-Reuse Coupling — arXiv:2609.22115},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.22115},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9536c141d6fa; sha256 5d3f0c1fd3e9ee68442c9a6fca7bf3b3916ead26279ac52d9ea90055c967308e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22115. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Zeroth-order optimization (ZOO) estimates updates from function evaluations, making perturbation queries a primary cost. Fixed budgets spend the same number of queries at every step, while adaptive controllers may offset their savings by using additional oracle calls to test estimator reliability. We introduce ZoAQ, an adaptive ZOO method built around query reuse. Rather than discarding past evaluations after each step, ZoAQ makes them useful for both the next update and the decision to query further. This enables adaptive query allocation without extra validation queries. Our analysis characterizes when this agreement identifies an update that supports descent and guides the controller to a sufficient query budget. On synthetic objectives, ZoAQ reduces queries by 43-48% relative to fixed baselines using 1.2M queries. In black-box attacks, it reaches 100% success with 320 and 625 average queries on MNIST and CIFAR-10, respectively. Across four OPT fine-tuning settings, ZoAQ saves 43-46% forward evaluations relative to fixed K=4, with accuracy changes within tasks ranging from -0.018 to +0.010. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.22115 — 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 2026-10-09T06:58:09.115Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).