提出新方法解决标签噪声中的语义污染问题，提升模型鲁棒性
Combating Semantic Contamination in Learning with Label Noise
- 通过跨视图与跨模型协作学习，从特征中提取合理语义关联
- 在合成与真实噪声数据集上均优于现有方法，显著降低语义污染影响
- 适合处理标签不干净的场景，尤其对鲁棒学习有帮助
标签噪声会损害深度神经网络的性能。现有标签修复方法常引入语义污染问题，即修复后的标签存在错误的语义关联。通过对代表性方法Robust LR的分析发现，使用视图的logits进行修复无法平衡各类别语义信息，而使用模型的logits又难以保持跨模型语义一致性。为此，本文提出协同交叉学习（Collaborative Cross Learning）方法，利用修复标签上的半监督学习，从多视图与多模型的嵌入中提取恰当的语义关系。实验表明，该方法在合成与真实噪声数据集上均优于现有方法，有效缓解了标签噪声和语义污染的影响。
原文摘要 · Abstract (English)
Noisy labels can negatively impact the performance of deep neural networks. One common solution is label refurbishment, which involves reconstructing noisy labels through predictions and distributions. However, these methods may introduce problematic semantic associations, a phenomenon that we identify as Semantic Contamination. Through an analysis of Robust LR, a representative label refurbishment method, we found that utilizing the logits of views for refurbishment does not adequately balance the semantic information of individual classes. Conversely, using the logits of models fails to maintain consistent semantic relationships across models, which explains why label refurbishment methods frequently encounter issues related to Semantic Contamination. To address this issue, we propose a novel method called Collaborative Cross Learning, which utilizes semi-supervised learning on refurbished labels to extract appropriate semantic associations from embeddings across views and models. Experimental results show that our method outperforms existing approaches on both synthetic and real-world noisy datasets, effectively mitigating the impact of label noise and Semantic Contamination.
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。