Matched but Not Better: An Equivalence-Tested Benchmark of Quantum Machine Learning on Six Clinical Datasets and the QML-Health-READY Reporting Framework
Computer Science > Machine Learning arXiv:2611.05442 (cs) [Submitted on 3 Nov 2026] Title: Matched but Not Better: An Equivalence-Tested Benchmark of Quantum Machine Learning on Six Clinical Datasets and the QML-Health-READY Reporting Framework Authors: Mohammed Siraj B, Gousia Habib, Ruban S.
计算机科学 > 机器学习 arXiv:2611.05442 (cs) [提交于 2026 年 11 月 3 日] 标题：匹配但并未更优：六个临床数据集上量子机器学习的等效性测试基准及 QML-Health-READY 报告框架。作者：Mohammed Siraj B, Gousia Habib, Ruban S。
Abstract: Background: Quantum machine learning (QML) is promoted for clinical prediction, but comparisons with classical models are rarely paired or tested for equivalence. Methods: We compared a quantum-kernel support vector machine (QSVM; 4 qubits, tuned bandwidth) and a variational quantum classifier (VQC) with five tuned classical models across six public clinical datasets. All models used identical four-component inputs and 25 cross-validation splits. Paired AUROC differences were assessed using a corrected resampled t-test, Holm-adjusted equivalence tests (margins: +/-0.02 and +/-0.05), and random-effects pooling. Ablation, noise, finite-shot, and implementation-fault analyses were conducted. We propose QML-Health-READY, an eight-domain reporting framework.
摘要：背景：量子机器学习（QML）被推广用于临床预测，但与经典模型的比较很少进行配对或等效性测试。方法：我们在六个公开临床数据集上，将量子核支持向量机（QSVM；4 量子比特，已调带宽）和变分量子分类器（VQC）与五个已调经典模型进行了比较。所有模型均使用相同的四分量输入和 25 次交叉验证拆分。配对 AUROC 差异通过修正重采样 t 检验、Holm 调整等效性检验（边界：+/-0.02 和 +/-0.05）以及随机效应合并进行评估。此外，还进行了消融、噪声、有限采样和实现故障分析。我们提出了 QML-Health-READY，这是一个包含八个领域的报告框架。
Results: Neither quantum model significantly outperformed any classical comparator. Pooled across datasets, QSVM differed from the kernel-matched radial basis function (RBF) SVM by -0.006 AUROC (95% CI: -0.017 to +0.004), equivalent to +/-0.02. VQC differed from the best classical model by -0.007 (95% CI: -0.019 to +0.004). QSVM was equivalent to RBF SVM within +/-0.05 in four of six datasets but was significantly worse than the best classical model on NHANES 2017-18 (-0.076). A classically trivial product-state kernel performed as well as the entangling kernel. With depolarizing noise (p = 0.01) and 1,024 shots per kernel entry, QSVM AUROC decreased by 0.054 (95% CI: 0.036 to 0.071). Untuned bandwidth produced near-chance AUROC (0.48-0.59).
结果：没有任何量子模型在性能上显著优于任何经典对比模型。在跨数据集汇总后，QSVM 与核匹配的径向基函数（RBF）SVM 相比，AUROC 差异为 -0.006（95% 置信区间：-0.017 至 +0.004），等效于 +/-0.02。VQC 与最佳经典模型相比差异为 -0.007（95% 置信区间：-0.019 至 +0.004）。在六个数据集中的四个里，QSVM 在 +/-0.05 范围内与 RBF SVM 等效，但在 NHANES 2017-18 数据集上显著劣于最佳经典模型（-0.076）。一种经典上平凡的积态核表现得与纠缠核一样好。在去极化噪声（p = 0.01）和每个核条目 1,024 次采样的情况下，QSVM 的 AUROC 下降了 0.054（95% 置信区间：0.036 至 0.071）。未调带宽产生的 AUROC 接近随机水平（0.48-0.59）。
Conclusions: Correctly configured quantum classifiers approached but did not exceed tuned classical models. Finite-shot estimation, rather than gate noise, was their main practical cost, while configuration errors could reverse conclusions. QML-Health-READY defines reporting requirements for credible claims.
结论：配置正确的量子分类器接近但并未超过已调的经典模型。有限采样估计而非门噪声是其主要的实际成本，而配置错误可能会导致结论反转。QML-Health-READY 为可信声明定义了报告要求。