动态发现全局假负样本，提升自监督对比学习效果
Discovering Global False Negatives On the Fly for Self-supervised Contrastive Learning
- 基于优化方法在线学习每个样本的阈值，自动识别语义相似的假负样本
- 在图像和图文数据上显著提升模型性能，优于传统负样本构造方式
- 计算开销与数据集规模无关，适合大规模训练场景
在自监督对比学习中，负样本通常通过锚点图像与除自身外整个数据集中的随机样本构建。然而，这种方法可能产生语义相似的负样本对，即“假负样本”，导致其嵌入被错误地拉远。为此，本文提出GloFND，一种基于优化的在线学习方法，可自动为每个锚点样本确定阈值，以在训练过程中全局识别假负样本。与以往局部检测方法不同，GloFND在整个数据集范围内发现假负样本，且每轮迭代的计算成本不随数据集规模增长。在图像和图文数据上的实验表明该方法有效。代码已开源：https://github.com/vibalcam/GloFND。
原文摘要 · Abstract (English)
In self-supervised contrastive learning, negative pairs are typically constructed using an anchor image and a sample drawn from the entire dataset, excluding the anchor. However, this approach can result in the creation of negative pairs with similar semantics, referred to as "false negatives", leading to their embeddings being falsely pushed apart. To address this issue, we introduce GloFND, an optimization-based approach that automatically learns on the fly the threshold for each anchor data to identify its false negatives during training. In contrast to previous methods for false negative discovery, our approach globally detects false negatives across the entire dataset rather than locally within the mini-batch. Moreover, its per-iteration computation cost remains independent of the dataset size. Experimental results on image and image-text data demonstrate the effectiveness of the proposed method. Our implementation is available at https://github.com/vibalcam/GloFND.
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。