首个利用复数相似度的量子自注意力模型，提升量子机器学习表达力。
Quantum Complex-Valued Self-Attention Model
- 提出复数线性酉组合框架，实现量子态间复数权重精确计算
- 4量子比特下在MNIST和Fashion-MNIST上分别达100%和99.2%准确率
- 适合关注量子注意力机制与复杂度建模的研究者
自注意力已革新经典机器学习，但现有量子自注意力模型因机制简化或不完整，未能充分发挥量子态潜力。为此，我们提出首个利用复数相似度的量子自注意力模型（QCSAM），通过将线性酉组合扩展为复数线性酉组合（CLCUs）框架，实现量子态间振幅与相位关系的更全面捕捉，并支持量子多头注意力。在MNIST和Fashion-MNIST上的实验表明，仅用4个量子比特，QCSAM即达到100%和99.2%的测试准确率，优于近期量子自注意力模型（如QKSAN、QSAN、GQHAN）。我们进一步评估了3-8量子比特及2-4类任务下的可扩展性，并通过消融实验证明复数权重优于实数权重。该工作通过增强量子自注意力的表达力与精度，推动了与量子力学本质一致的量子机器学习发展。
原文摘要 · Abstract (English)
Self-attention has revolutionized classical machine learning, yet existing quantum self-attention models underutilize quantum states' potential due to oversimplified or incomplete mechanisms. To address this limitation, we introduce the Quantum Complex-Valued Self-Attention Model (QCSAM), the first framework to leverage complex-valued similarities, which captures amplitude and phase relationships between quantum states more comprehensively. To achieve this, QCSAM extends the Linear Combination of Unitaries (LCUs) into the Complex LCUs (CLCUs) framework, enabling precise complex-valued weighting of quantum states and supporting quantum multi-head attention. Experiments on MNIST and Fashion-MNIST show that QCSAM outperforms recent quantum self-attention models, including QKSAN, QSAN, and GQHAN. With only 4 qubits, QCSAM achieves 100% and 99.2% test accuracies on MNIST and Fashion-MNIST, respectively. Furthermore, we evaluate scalability across 3-8 qubits and 2-4 class tasks, while ablation studies validate the advantages of complex-valued attention weights over real-valued alternatives. This work advances quantum machine learning by enhancing the expressiveness and precision of quantum self-attention in a way that aligns with the inherent complexity of quantum mechanics.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。