Deep neural networks (DNNs) have achieved remarkable success in a variety of computer vision tasks, where massive labeled images are routinely required for model optimization. Yet, the data collected from the open world are unavoidably polluted by noise, which may significantly undermine the efficacy of the learned models. Various attempts have been made to reliably train DNNs under data noise, but they separately account for either the noise existing in the labels or that existing in the images. A naive combination of the two lines of works would suffer from the limitations in both sides, and miss the opportunities to handle the two kinds of noise in parallel. This work provides a first, unified framework for reliable learning under the joint (image, label)-noise. Technically, we develop a confidence-based sample filter to progressively filter out noisy data without the need of pre-specifying noise ratio. Then, we penalize the model uncertainty of the detected noisy data instead of letting the model continue over-fitting the misleading information in them. Experimental results on various challenging synthetic and real-world noisy datasets verify that the proposed method can outperform competing baselines in the aspect of classification performance.

翻译：深度神经网络（DNNs）在各种计算机视觉任务中取得了显著成功，这些任务通常需要大量标注图像来进行模型优化。然而，从开放世界收集的数据不可避免地会受到噪声污染，这可能严重削弱所学模型的有效性。已有多种尝试旨在可靠地训练含数据噪声的DNNs，但它们分别只考虑标签中存在的噪声或图像中存在的噪声。简单地将这两类方法进行组合会受限于双方的不足，并错失并行处理这两种噪声的机会。本研究首次提出了一个统一的框架，用于在联合的（图像、标签）噪声下进行可靠学习。技术上，我们开发了一个基于置信度的样本过滤器，能够逐步过滤掉噪声数据，而无需预先指定噪声比例。接着，我们对检测到的噪声数据的模型不确定性进行惩罚，而不是让模型继续过拟合其中包含的误导信息。在多种具有挑战性的合成和真实世界含噪声数据集上的实验结果表明，所提方法在分类性能方面优于竞争基线。