arXiv 2505.220282025-05-28cs.LG

面向不精确类标签的弱监督对比学习

Weakly-Supervised Contrastive Learning for Imprecise Class Labels

School of Computer Science and Engineering, Southeast University, Nanjing, China（计算机科学与工程学院，东南大学，南京，中国）；Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, China（计算机网络与信息集成重点实验室（东南大学），教育部，中国）

Zi-Hao Zhou, Jun-Jie Wang, Tong Wei, Min-Ling Zhang

备注：: 38 pages, 2 figures, 11 tables

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对不精确类标签限制监督对比学习的问题，提出基于连续语义相似度的图论弱监督对比学习框架，通过迭代细化弱监督信号测量样本对相似度，在噪声和部分标签场景下显著提升性能并具备理论误差保证。

摘要：

对比学习在学习有效表示方面取得了显著成功，其中监督对比学习通常优于自监督方法。然而在现实场景中，数据标注往往具有模糊性或不准确性，这意味着类标签可能无法可靠地指示两个样本是否属于同一类别。这一局限性限制了监督对比学习的适用性。为应对此挑战，我们引入“连续语义相似度”的概念来定义正负样本对。我们不直接依赖不精确的类标签，而是测量样本对之间的语义相似度，通过迭代细化弱监督信号来量化它们属于同一类别的紧密程度。基于此概念，我们提出了一个用于弱监督对比学习的图论框架，其中语义相似度作为图权重。我们的框架具有高度通用性，可应用于许多弱监督学习场景。我们通过在两种常见设置（即噪声标签和部分标签学习）中的实验证明了其有效性，在这些设置中，现有方法可以轻松集成以显著提升性能。在理论上，我们为该方法建立了误差界限，表明在温和条件下它可以逼近监督对比学习。实现代码可在 https://github.com/Speechless-10308/WSC 获取。

英文摘要：

Contrastive learning has achieved remarkable success in learning effective representations, with supervised contrastive learning often outperforming self-supervised approaches. However, in real-world scenarios, data annotations are often ambiguous or inaccurate, meaning that class labels may not reliably indicate whether two examples belong to the same class. This limitation restricts the applicability of supervised contrastive learning. To address this challenge, we introduce the concept of ``continuous semantic similarity'' to define positive and negative pairs. Instead of directly relying on imprecise class labels, we measure the semantic similarity between example pairs, which quantifies how closely they belong to the same category by iteratively refining weak supervisory signals. Based on this concept, we propose a graph-theoretic framework for weakly-supervised contrastive learning, where semantic similarity serves as the graph weights. Our framework is highly versatile and can be applied to many weakly-supervised learning scenarios. We demonstrate its effectiveness through experiments in two common settings, i.e., noisy label and partial label learning, where existing methods can be easily integrated to significantly improve performance. Theoretically, we establish an error bound for our approach, showing that it can approximate supervised contrastive learning under mild conditions. The implementation code is available at https://github.com/Speechless-10308/WSC.

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。