arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.05913cs.LGstat.ML
收藏

少可以多吗?当增加到平衡标签噪声率被认为是有益的时

Can Less be More? When Increasing-to-Balancing Label Noise Rates Considered Beneficial

University of California, Santa Cruz(加州大学圣克鲁兹分校)

Yang Liu, Jialu Wang

摘要:

在本文中,我们回答了插入标签噪声(信息量较少的标签)何时可以返回更准确和公平的模型。我们主要受到三个观察结果的启发:1)与降低标签噪声率相比,增加噪声率更容易实现; 2)增加某类实例的标签噪声以平衡噪声率(增加到平衡)导致更容易的学习问题; 3)增加到平衡提高了对标签偏见的公平性保证。在本文中,我们首先量化的权衡引入增加一定组的实例的标签噪声率w.r.t.标签信息量的损失和学习难度的降低。我们分析表明,当这样的增加是有益的,无论是提高泛化能力或公平性保证。然后,我们提出了一种方法来插入标签噪声正确的学习任务与嘈杂的标签,无论是没有或有公平性约束。我们面临的主要技术挑战是由于我们不知道哪些数据实例受到更高噪声的影响,并且我们没有地面真值标签来验证任何可能的假设。我们提出了一种检测方法,该方法可以在不使用地面真实标签的情况下通知我们哪组标签可能会遭受更高的噪声。我们正式建立所提出的解决方案的有效性,并证明了它与广泛的实验。

英文摘要:

In this paper, we answer the question of when inserting label noise (less informative labels) can instead return us more accurate and fair models. We are primarily inspired by three observations: 1) In contrast to reducing label noise rates, increasing the noise rates is easy to implement; 2) Increasing a certain class of instances' label noise to balance the noise rates (increasing-to-balancing) results in an easier learning problem; 3) Increasing-to-balancing improves fairness guarantees against label bias. In this paper, we first quantify the trade-offs introduced by increasing a certain group of instances' label noise rate w.r.t. the loss of label informativeness and the lowered learning difficulties. We analytically demonstrate when such an increase is beneficial, in terms of either improved generalization power or the fairness guarantees. Then we present a method to insert label noise properly for the task of learning with noisy labels, either without or with a fairness constraint. The primary technical challenge we face is due to the fact that we would not know which data instances are suffering from higher noise, and we would not have the ground truth labels to verify any possible hypothesis. We propose a detection method that informs us which group of labels might suffer from higher noise without using ground truth labels. We formally establish the effectiveness of the proposed solution and demonstrate it with extensive experiments.

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。

↑