arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2001.03772cs.LGstat.ML
收藏

置信度使依赖于实例的标签噪声学习成为可能

Confidence Scores Make Instance-dependent Label-noise Learning Possible

RIKEN Center for Advanced Intelligence Project(理化学研究所先进智能项目中心);Hong Kong Baptist University(香港浸会大学);University of Sydney(悉尼大学);University of Tokyo(东京大学)

Antonin Berthon, Bo Han, Gang Niu, Tongliang Liu, Masashi Sugiyama

摘要:

在带噪声标签的学习中,对于每个实例,其标签可以随机地沿着称为噪声模型的过渡分布移动到其他类。研究得很好的噪声模型都是与实例无关的,也就是说,转换只取决于原始标签,而不是实例本身,因此它们在野外不太实用。幸运的是,已经研究了基于实例相关噪声的方法,但大多数方法都依赖于对噪声模型的强假设。为了缓解这个问题,我们引入了置信度评分的实例相关噪声(CSIDN),其中每个实例标签对都配备了置信度评分。我们发现,在置信度的帮助下,每个实例的转移分布可以近似估计。类似于强大的前向校正实例无关的噪声,我们提出了一种新的CSIDN的实例级前向校正。我们证明了我们的方法的实用性和有效性,通过多个实验下的合成标签噪声和真实世界的未知噪声。

英文摘要:

In learning with noisy labels, for every instance, its label can randomly walk to other classes following a transition distribution which is named a noise model. Well-studied noise models are all instance-independent, namely, the transition depends only on the original label but not the instance itself, and thus they are less practical in the wild. Fortunately, methods based on instance-dependent noise have been studied, but most of them have to rely on strong assumptions on the noise models. To alleviate this issue, we introduce confidence-scored instance-dependent noise (CSIDN), where each instance-label pair is equipped with a confidence score. We find with the help of confidence scores, the transition distribution of each instance can be approximately estimated. Similarly to the powerful forward correction for instance-independent noise, we propose a novel instance-level forward correction for CSIDN. We demonstrate the utility and effectiveness of our method through multiple experiments under synthetic label noise and real-world unknown noise.

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。

↑