量子概率标签精炼：提升标签质量以实现鲁棒图像分类
Quantum Probabilistic Label Refining: Enhancing Label Quality for Robust Image Classification
AI总结：
针对传统独热标签训练导致的过自信与鲁棒性差问题，提出量子概率标签精炼混合框架，用量子电路生成含输入特定不确定性的软标签训练CNN，显著提升噪声下准确率与模型校准性。
摘要：
基于独热标签使用softmax交叉熵进行学习，往往会导致预测过于自信，且在噪声或扰动下鲁棒性较差。标签平滑通过均匀重新分配部分置信度来缓解这一问题，但它对所有样本一视同仁，忽略了类内变异性。我们提出一种混合量子-经典框架，利用量子非确定性将数据标签精炼为概率标签，相比标签平滑或贝叶斯方法，能提供更细致、类人化的不确定性表征。变分量子电路（VQC）将输入编码为多量子比特量子态，利用纠缠和叠加捕捉细微的特征相关性。通过玻恩规则进行测量，提取出能反映输入特定不确定性的概率软标签。随后，这些标签被用于训练采用软目标交叉熵损失的经典卷积神经网络（CNN）。在MNIST和Fashion-MNIST数据集上，我们的方法提升了鲁棒性：在噪声条件下，准确率最高提升50%，同时在干净数据上保持了具有竞争力的准确率。该方法还增强了模型的校准性和可解释性，因为CNN的输出能更好地反映量子衍生的不确定性。本研究提出了量子概率标签精炼（Quantum Probabilistic Label Refining）方法，通过精炼的、具备相关性感知能力的标签，在无需架构改动或对抗技术的情况下，搭建起量子测量与经典深度学习之间的桥梁，用于鲁棒训练。
英文摘要：
Learning with softmax cross-entropy on one-hot labels often leads to overconfident predictions and poor robustness under noise or perturbations. Label smoothing mitigates this by redistributing some confidence uniformly, but treats all samples equally, ignoring intra-class variability. We propose a hybrid quantum-classical framework that leverages quantum non-determinism to refine data labels into probabilistic ones, offering more nuanced, human-like uncertainty representations than label smoothing or Bayesian approaches. A variational quantum circuit (VQC) encodes inputs into multi-qubit quantum states, using entanglement and superposition to capture subtle feature correlations. Measurement via the Born rule extracts probabilistic soft labels that reflect input-specific uncertainty. These labels are then used to train a classical convolutional neural network (CNN) with soft-target cross-entropy loss. On MNIST and Fashion-MNIST, our method improves robustness, achieving up to 50% higher accuracy under noise while maintaining competitive accuracy on clean data. It also enhances model calibration and interpretability, as CNN outputs better reflect quantum-derived uncertainty. This work introduces Quantum Probabilistic Label Refining, bridging quantum measurement and classical deep learning for robust training via refined, correlation-aware labels without architectural changes or adversarial techniques.
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。
