arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.14070quant-phcs.AI
收藏

量子费舍尔有效维度能否评判混合量子神经网络?

What Does the Effective Dimension of a Hybrid Quantum Neural Network Actually Measure?

Qatar University(卡塔尔大学)

Md Masum Mia, Amith Khandakar, Mohamed Arselene Ayari

AI总结:

该研究验证了量子费舍尔有效维度的实际含义,发现其表征电路几何而非模型质量,还提出了对应诊断问题的实用方案。

摘要:

混合量子神经网络(QNN)将经典特征提取器与变分量子电路相结合,电路量子费舍尔信息矩阵(QFIM)的有效维度被广泛用于评估这类模型。该标量常被解释为可训练性或准确率的指标,但这类解读很少在受控条件下得到验证。因此我们研究它实际测量的是什么:将有效维度定义为QFIM对角元的参与率,取值介于1和参数数量之间;在两个图像数据集、四个冻结骨干网络(含随机投影对照)、8量子比特上的四种电路深度、三个随机种子的设置下进行评估。训练期间,有效维度始终远低于参数数量,其与测试准确率的相关系数在-0.72至+0.87之间,半数配置下会随种子改变符号;增加深度会激活更多电路方向,占参数数量的比例从0.38升至0.91,而初始化梯度方差保持在1e-5至1e-4之间,因此8量子比特下未出现贫瘠高原(BP)。固定电路时,四个骨干网络的初始有效维度为6至9,尽管CIFAR-10的准确率在23.14%至86.44%之间。这些结果表明,有效维度表征的是电路几何而非模型质量;最后提出了将各诊断方法对应到可回答问题的实用方案。

英文摘要:

Hybrid quantum neural networks (QNNs) pass features from a classical backbone into a variational quantum circuit, and many studies log a scalar called the effective dimension as a sign of capacity or model quality. The effective dimension is usually defined as the participation ratio (PR) of the quantum Fisher information matrix (QFIM), but studies in practice log a PR of cheap parameter sensitivities instead, and the meaning of that scalar is rarely tested. We show that the logged scalar equals the PR of Pauli-Z output sensitivities, that each sensitivity is at most n times its QFIM diagonal entry on n qubits, and that a PR of the QFIM diagonal overstates the full-matrix PR by an exact factor set by the off-diagonal mass. We then test the scalar on CIFAR-10 and EuroSAT with four frozen backbones, one of them a random projection, over 4 to 16 qubits, 1 to 4 layers, three entangling patterns, and three seeds. At a fixed circuit, CIFAR-10 accuracy spans 23.14% to 86.44% while the scalar stays between 6.63 and 9.13, and its correlation with accuracy changes sign across seeds in 7 of the 14 configurations. Depth raises the share of active parameters from 0.39 to 0.91, near a ceiling we derive, and initial gradient variance falls by a factor of 34.5 from 4 to 16 qubits. Removing entanglement raises accuracy by 5.3 to 8.2 points at the same parameter count. The effective dimension therefore describes circuit geometry rather than model quality, so capacity, trainability, and accuracy each need their own diagnostic.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑