Masking: A New Perspective of Noisy Supervision
Masking: A New Perspective of Noisy Supervision
复制标题
DOI：
--
复制
发表时间：
2018-05
期刊：
影响因子：
--
通讯作者：
Bo Han;Jiangchao Yao;Gang Niu;Mingyuan Zhou;I. Tsang;Ya Zhang;Masashi Sugiyama
中科院分区：
文献类型：
--
作者：
Bo Han;Jiangchao Yao;Gang Niu;Mingyuan Zhou;I. Tsang;Ya Zhang;Masashi Sugiyama
重要的是要学习各种类型的分类器，给出带有噪声标签的训练数据。在迄今为止最流行的噪声模型中，噪声标签被未知的噪声转移矩阵从地面真实标签中破坏。因此，通过估计这个矩阵，分类器可以避免过度拟合那些嘈杂的标签。然而，由于两步方法的间接性质，或者没有足够大的数据来提供端到端的方法，这种估计实际上是困难的。在本文中，我们提出了一种称为掩蔽的人工辅助方法，它传达了人类对无效类转换的认知，并自然地推测了噪声转换矩阵的结构。为此，我们推导出一个结构感知的概率模型，结合结构先验，并解决了结构提取和结构对齐的挑战。由于掩蔽，我们只估计未掩蔽的噪声转移概率，估计的负担大大减少。我们在三种噪声结构的CIFAR-10和CIFAR-100以及不可知噪声结构的工业级Clothing 1 M上进行了大量的实验，结果表明掩蔽可以显著提高分类器的鲁棒性。
It is important to learn various types of classifiers given training data with noisy labels. Noisy labels, in the most popular noise model hitherto, are corrupted from ground-truth labels by an unknown noise transition matrix. Thus, by estimating this matrix, classifiers can escape from overfitting those noisy labels. However, such estimation is practically difficult, due to either the indirect nature of two-step approaches, or not big enough data to afford end-to-end approaches. In this paper, we propose a human-assisted approach called Masking that conveys human cognition of invalid class transitions and naturally speculates the structure of the noise transition matrix. To this end, we derive a structure-aware probabilistic model incorporating a structure prior, and solve the challenges from structure extraction and structure alignment. Thanks to Masking, we only estimate unmasked noise transition probabilities and the burden of estimation is tremendously reduced. We conduct extensive experiments on CIFAR-10 and CIFAR-100 with three noise structures as well as the industrial-level Clothing1M with agnostic noise structure, and the results show that Masking can improve the robustness of classifiers significantly.