Discover and Mitigate Unknown Biases with Debiasing Alternate Networks-猫眼课题宝

DOI：

10.48550/arxiv.2207.10077

发表时间：

2022-07

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Zhiheng Li;A. Hoogs;Chenliang Xu

人们发现深度图像分类器可以从数据集中学习偏差。为了减轻偏差，大多数以前的方法都需要受保护属性（例如年龄、肤色）的标签作为完全监督，这有两个限制：1）当标签不可用时，这是不可行的； 2）它们无法减轻未知的偏见——人类不会先入为主的偏见。为了解决这些问题，我们提出了去偏替代网络（DebiAN），它包含两个网络——发现器和分类器。通过以替代方式进行训练，发现者尝试在没有任何偏差注释的情况下找到分类器的多个未知偏差，并且分类器旨在忘却发现者识别的偏差。虽然之前的工作是根据单个偏差来评估去偏差结果，但我们创建了多色 MNIST 数据集，以更好地对多偏差设置中多个偏差的缓解进行基准测试，这不仅揭示了以前方法中的问题，而且还展示了 Debian 在同时识别和减轻多个偏差方面的优势。我们进一步对现实世界的数据集进行了广泛的实验，表明 Debian 的发现者可以识别人类可能难以发现的未知偏差。在去偏差方面，Debian 实现了强大的偏差缓解性能。

Deep image classifiers have been found to learn biases from datasets. To mitigate the biases, most previous methods require labels of protected attributes (e.g., age, skin tone) as full-supervision, which has two limitations: 1) it is infeasible when the labels are unavailable; 2) they are incapable of mitigating unknown biases -- biases that humans do not preconceive. To resolve those problems, we propose Debiasing Alternate Networks (DebiAN), which comprises two networks -- a Discoverer and a Classifier. By training in an alternate manner, the discoverer tries to find multiple unknown biases of the classifier without any annotations of biases, and the classifier aims at unlearning the biases identified by the discoverer. While previous works evaluate debiasing results in terms of a single bias, we create Multi-Color MNIST dataset to better benchmark mitigation of multiple biases in a multi-bias setting, which not only reveals the problems in previous methods but also demonstrates the advantage of DebiAN in identifying and mitigating multiple biases simultaneously. We further conduct extensive experiments on real-world datasets, showing that the discoverer in DebiAN can identify unknown biases that may be hard to be found by humans. Regarding debiasing, DebiAN achieves strong bias mitigation performance.

Discover and Mitigate Unknown Biases with Debiasing Alternate Networks

文献摘要

相似文献