利用通用代理模型进行样本选择以对抗标签噪声
Combating Label Noise With A General Surrogate Model For Sample Selection
AI总结：
本文提出利用免训练的视觉-语言代理模型CLIP自动筛选干净样本，并设计边际自适应损失校正选择偏差，从而在无需CLIP参与推理的情况下有效对抗标签噪声。
摘要：
现代深度学习系统对数据需求巨大。利用网络数据进行学习是可行的解决方案之一，但不可避免地会引入标签噪声，从而损害深度神经网络的性能。样本选择是处理标签噪声的有效方法，其关键在于依据某种准则分离干净样本。以往方法更关注小损失准则，即将小损失样本视为干净样本。然而，这种策略依赖于每个数据实例的学习动态。由于频繁出现的损坏学习模式，一些噪声样本仍会被记忆。为解决这一问题，免训练的代理模型更受青睐，可避免记忆效应的影响。在本工作中，我们提出利用视觉-语言代理模型CLIP自动过滤噪声样本。CLIP通过其文本-图像对齐能力引入外部知识，以促进干净样本的选择。此外，我们设计了一种边际自适应损失来正则化由CLIP引入的选择偏差，从而提供对标签噪声的鲁棒性。我们在真实世界和合成噪声数据集上验证了所提方法的有效性。在推理阶段不涉及CLIP的情况下，我们的方法取得了显著改进。
英文摘要：
Modern deep learning systems are data-hungry. Learning with web data is one of the feasible solutions, but will introduce label noise inevitably, which can hinder the performance of deep neural networks. Sample selection is an effective way to deal with label noise. The key is to separate clean samples based on some criterion. Previous methods pay more attention to the small loss criterion where small-loss samples are regarded as clean ones. Nevertheless, such a strategy relies on the learning dynamics of each data instance. Some noisy samples are still memorized due to frequently occurring corrupted learning patterns. To tackle this problem, a training-free surrogate model is preferred, freeing from the effect of memorization. In this work, we propose to leverage the vision-language surrogate model CLIP to filter noisy samples automatically. CLIP brings external knowledge to facilitate the selection of clean samples with its ability of text-image alignment. Furthermore, a margin adaptive loss is designed to regularize the selection bias introduced by CLIP, providing robustness to label noise. We validate the effectiveness of our proposed method on both real-world and synthetic noisy datasets. Our method achieves significant improvement without CLIP involved during the inference stage.
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。
