提出可自适应优化阈值的半监督学习方法，让模型自动调节伪标签筛选标准。
MTSSL: Meta-Thresholding Semi-Supervised Learning
- 将阈值τ视为可微参数，通过元优化动态调整伪标签筛选标准
- 实验显示不同阈值下模型性能曲线重合，验证理论有效性
- 适合追求简化调参、提升鲁棒性的半监督学习研究者
大量半监督学习（SSL）算法依赖阈值τ来选择伪标签。尽管不同算法中τ的取值因学习视角而异，但性能却相近。这促使我们建立统一的理论框架，解释τ在SSL中的作用。我们从统计角度证明，无监督损失受正确与错误伪标签独立影响，而τ通过调节两者数量来平衡相应误差项。这种内在权衡表明，相同损失可通过不同τ实现，训练过程中精确最优τ可能非必需。基于此，我们将τ视为可更新参数，通过微分进行优化，提出新策略——元阈值半监督学习（MTSSL）。大量实验表明MTSSL性能更优。我们观察到，即使τ值差异显著，不同算法的准确率曲线仍可完全重合，支持了理论框架，并提示未来设计可放宽对τ的选择要求。
原文摘要 · Abstract (English)
A large body of Semi-supervised Learning~(SSL) algorithms encounter the threshold $τ$ to select pseudo-labels. The value of $τ$ across different SSL algorithms can vary depending on the learning perspective, yet they may achieve similar performance. It motivates us to establish a unified theoretical framework to explain the role of $τ$ in SSL. We statistically explained that the unsupervised loss is affected independently by correct and incorrect pseudo-labels, while $τ$ adjusts their numbers to balance the corresponding error term. This inherent trade-off indicates that SSL can reach the same loss with varying $τ$, precise optimal values of $τ$ during training may be unnecessary. With this, we treat $τ$ as an updatable parameter and optimize it via differentiation; the new policy is named \textbf{Meta-Thresholding Semi-Supervised Learning (MTSSL)}. Extensive experiments demonstrate the superior performance of MTSSL. We observe that the accuracy curves of SSL algorithms can overlap completely even when the values of $τ$ differ significantly, which supports our theoretical framework and indicates that the selection of $τ$ can be relaxed in the future design of SSL algorithms.
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。