通过熵加权提升伪标签质量，增强半监督对比学习效果
Improving Semi-Supervised Contrastive Learning via Entropy-Weighted Confidence Integration of Anchor-Positive Pairs
- 用预测分布熵衡量样本置信度，动态加权伪标签
- 在低标注数据下仍保持高准确率与稳定训练
- 适合标签稀缺场景下的图像分类任务
传统半监督对比学习仅对最高预测概率超过阈值的样本分配伪标签，并基于这些样本进行有监督对比学习。本文提出一种新型损失函数，根据预测概率分布的熵估计样本置信度，并施加基于置信度的自适应权重。该方法使原本被排除训练的样本也能获得伪标签，实现了对锚点与正样本置信度更合理地联合建模的对比学习。实验表明，该方法显著提升分类准确率，在低标注条件下表现更稳定。
原文摘要 · Abstract (English)
Conventional semi-supervised contrastive learning methods assign pseudo-labels only to samples whose highest predicted class probability exceeds a predefined threshold, and then perform supervised contrastive learning using those selected samples. In this study, we propose a novel loss function that estimates the confidence of each sample based on the entropy of its predicted probability distribution and applies confidence-based adaptive weighting. This approach enables pseudo-label assignment even to samples that were previously excluded from training and facilitates contrastive learning that accounts for the confidence of both anchor and positive samples in a more principled manner. Experimental results demonstrate that the proposed method improves classification accuracy and achieves more stable learning performance even under low-label conditions.
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。