arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2202.02016cs.LGstat.ML
收藏

标签噪声转移矩阵的可辨识性

Identifiability of Label Noise Transition Matrix

University of California, Santa Cruz(加州大学圣克鲁兹分校);Carnegie Mellon University(卡内基梅隆大学);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)

Yang Liu, Hao Cheng, Kun Zhang

摘要:

噪声转移矩阵在带噪声标签的学习问题中起着核心作用。在许多其他原因中,大量现有的解决方案依赖于对它的访问。识别和估计没有地面真值标签的转移矩阵是一项关键且具有挑战性的任务。当标签噪声转移依赖于每个实例时,识别实例依赖的噪声转移矩阵的问题变得更具挑战性。尽管最近的工作提出了解决方案,从实例相关的嘈杂标签学习,该领域缺乏统一的理解,当这样的问题仍然是可识别的。本文的目的是刻画标签噪声转移矩阵的可识别性。在Kruskal的可识别性结果的基础上,我们能够证明多个噪声标签在识别实例级别的通用情况下的噪声转移矩阵的必要性。我们进一步实例化的结果,以解释国家的最先进的解决方案的成功,以及如何额外的假设减轻了多个嘈杂的标签的要求。我们的研究结果还表明,解开的功能是有助于在上述识别任务,我们提供了经验证据。

英文摘要:

The noise transition matrix plays a central role in the problem of learning with noisy labels. Among many other reasons, a large number of existing solutions rely on access to it. Identifying and estimating the transition matrix without ground truth labels is a critical and challenging task. When label noise transition depends on each instance, the problem of identifying the instance-dependent noise transition matrix becomes substantially more challenging. Despite recent works proposing solutions for learning from instance-dependent noisy labels, the field lacks a unified understanding of when such a problem remains identifiable. The goal of this paper is to characterize the identifiability of the label noise transition matrix. Building on Kruskal's identifiability results, we are able to show the necessity of multiple noisy labels in identifying the noise transition matrix for the generic case at the instance level. We further instantiate the results to explain the successes of the state-of-the-art solutions and how additional assumptions alleviated the requirement of multiple noisy labels. Our result also reveals that disentangled features are helpful in the above identification task and we provide empirical evidence.

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。

↑