一句话：提出 RULER，通过区分敏感/非敏感属性生成修复用例，并以两阶段迭代对抗训练提升 DNN 公平性。
深度神经网络在自主驾驶、金融管理等应用中面临决策伦理问题，公平性尤需关注。已有公平性测试技术通过生成个体歧视实例等测试用例来修复模型，但许多用例对提升公平性并不直接有效，且带来大量计算开销。RULER 在测试用例生成阶段区分敏感属性与非敏感属性，将生成的有效用例用于模型训练；同时把训练过程分解为两个阶段，并引入一种新的迭代对抗训练方法来平衡准确率与公平性。在四个数据集上与当前最佳技术相比，RULER 生成的有效修复测试用例多 7–28 倍，测试生成快 10–15 倍，公平性提升平均多 26–43%。
Deep Neural Networks (DNNs) are becoming an integral part of many real-world applications, such as autonomous driving and financial management. While these models enable autonomy, there are however concerns regarding their ethics in decision making. For instance, fairness is an aspect that requires particular attention. A number of fairness testing techniques have been proposed to address this issue, e.g., by generating test cases called individual discriminatory instances for repairing DNNs. Although they have demonstrated great potential, they tend to generate many test cases that are not directly effective in improving fairness and incur substantial computation overhead. We propose a new model repair technique, RULER, by discriminating sensitive and non-sensitive attributes during test case generation for model repair. The generated cases are then used in training to improve DNN fairness. RULER balances the trade-off between accuracy and fairness by decomposing the training procedure into two phases and introducing a novel iterative adversarial training method for fairness. Compared to the state-of-the-art techniques on four datasets, RULER has 7-28 times more effective repair test cases generated, is 10-15 times faster in test generation, and has 26-43% more fairness improvement on average.
DOI 原文 ·
@inproceedings{paperbot7125,
title = {RULER: discriminative and iterative adversarial training for deep neural network fairness},
author = {Guanhong Tao and Weisong Sun and Tingxu Han and Chunrong Fang and Xiangyu Zhang},
booktitle = {FSE},
year = {2022},
doi = {10.1145/3540250.3549169}
}