arXiv：cs.LG· Lu Wei, Yufeng Wang, Haibin Ling·· 4 小时前AI 评分27
报告惯例掩盖了什么：对精确计算度量下量子自然梯度的匹配预算审计
What a Reporting Convention Hides: A Matched-Budget Audit of Quantum Natural Gradient with an Exactly Computed Metric
AI 导读
一项匹配预算审计发现，只统计达到目标损失的运行会掩盖 SPSA 与 Adam 的真实差距：在电路宽度中位数上，SPSA 达到宽松目标所需评估次数是 Adam 的两倍以上，剔除未达标运行后这一差距消失。
正文
Abstract:Several published comparisons of variational quantum optimizers time only runs that reach a target loss, or read the verdict at a single target. Either convention could decide whether an optimizer's costlier steps pay off. We measure how much each convention changes verdicts among Adam, simultaneous perturbation stochastic approximation (SPSA) and quantum natural gradient (QNG), on initializations held out from the selection of settings. We compute the exact metric that preconditions QNG, price every step in circuit evaluations and give every method the same budget. In a median pooled over circuit widths, cost families and a sweep of settings with common misses, SPSA needs more than twice Adam's evaluations to reach a loose target. Dropping the censored runs that miss the target hides this gap. On the global-cost family we hold fixed the settings selected for a strict target. QNG then usually reaches the loose target after Adam but the strict target first. QNG's strict-target lead disappears when the metric's simulator price, linear in the number of parameters, is replaced by an assumed hardware count quadratic in that number. We recommend charging every miss the budget and reporting verdicts across targets.
| Comments: | 29 pages, 6 figures, 12 tables. Lu Wei and Yufeng Wang contributed equally |
| Subjects: | Machine Learning (cs.LG); Quantum Physics (quant-ph) |
| Cite as: | arXiv:2610.09425 [cs.LG] |
| (or arXiv:2610.09425v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09425 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lu Wei [view email]
[v1]
Wed, 7 Oct 2026 04:24:54 UTC (1,151 KB)
来源：arXiv：cs.LG · arxiv.org