用于跨模态分类的经典数据集上量子嵌入Transformer的研究

中文摘要 AI 辅助

我们测试参数化量子电路（PQC）是否能提升混合量子-经典模型在经典数据集上的性能，采用接口匹配的经典映射作为对照，同时固定所有其他组件。我们的架构Quantum-Embedded Attention（QEA，量子嵌入注意力机制）使用可学习投影器将骨干特征压缩为$n_q$维角度向量，用浅层PQC将这些角度映射到单量子比特和两量子比特泡利期望值，再通过经典注意力解码器生成类别logits。我们假设，与输入输出维度匹配的经典映射相比，PQC能提升准确率或种子间稳定性。我们在威斯康星乳腺癌数据集上进行了接口匹配的$2\times2$因子实验，$n_q$取值为$\boldsymbol{\boldsymbol{4}}$和$\boldsymbol{\boldsymbol{8}}$，每个单元独立将PQC替换为经典映射，将注意力解码器替换为线性头，每个配对种子有5个。4个配对量子-经典对比的95%置信区间中，有3个包含0；第4个是$n_q=4$时注意力解码器的+1.63个百分点对比，在$n_q=8$时符号反转，且无法在4个对比中通过校正。因此，实验显示PQC无一致贡献，无法确立等价性。5数据集跨模态网格实验显示，在AG News、威斯康星乳腺癌和BirdCLEF上准确率相当，但在CIFAR-10上存在较大差距；这些单元未接口匹配，仅作描述性解读。我们报告了所有计划的标准运行，区分了当前泡利读出结果与传统概率读出实验，并分析了瓶颈、模拟、有限样本和噪声限制。结果未确立量子优势，表明在将混合模型的性能归因于其量子层前，有必要进行受控组件归因。

英文摘要

We test whether a parameterized quantum circuit (PQC) improves a hybrid quantum-classical model's performance on classical datasets, using an interface-matched classical map as the control while holding all other components fixed. Our architecture, Quantum-Embedded Attention (QEA), uses a learnable projector to compress backbone features into an $n_q$-dimensional angle vector, a shallow PQC to map those angles to one- and two-qubit Pauli expectations, and a classical attention decoder to produce class logits. We hypothesized the PQC would improve accuracy or seed-to-seed stability over a classical map with matched input/output dimensions. We test this with an interface-matched $2\times2$ factorial on Breast Cancer Wisconsin at $n_q\in\{4,8\}$, independently swapping the PQC for a classical map and the attention decoder for a linear head, across five paired seeds per cell. Three of four paired quantum-minus-classical $95\%$ confidence intervals include zero; the fourth, a $+1.63$ percentage-point contrast for the attention decoder at $n_q=4$, reverses sign at $n_q=8$ and does not survive correction across the four contrasts. The experiment thus shows no consistent PQC contribution and cannot establish equivalence. A five-dataset cross-modality grid shows comparable accuracy on AG~News, Breast Cancer Wisconsin, and BirdCLEF but a large deficit on CIFAR-10; these cells are not interface-matched and are interpreted descriptively. We report all planned canonical runs, distinguish current Pauli-readout results from legacy probability-readout experiments, and analyze bottleneck, simulation, finite-shot, and noise limitations. The results do not establish a quantum advantage; they demonstrate why controlled component attribution is necessary before crediting a hybrid model's performance to its quantum layer.

发表机构

• University of London（伦敦大学）

机构由 AI 辅助整理，请以论文原文为准。