Efficient white-box fairness testing through gradient search
Efficient white-box fairness testing through gradient search
复制标题
DOI：
10.1145/3460319.3464820
复制
发表时间：
2021-07
期刊：
影响因子：
--
通讯作者：
Lingfeng Zhang;Yueling Zhang;M. Zhang
中科院分区：
文献类型：
--
作者：
Lingfeng Zhang;Yueling Zhang;M. Zhang
深度学习（DL）系统越来越多地应用于广泛的自主决策。除了鲁棒性和安全性之外，公平性也是一个设计良好的深度学习系统应该具备的重要属性。为了评估和提高模型的个体公平性，在输入空间中识别个体歧视性实例的系统测试用例生成是必不可少的。在本文中，我们提出了一个有效发现个人公平违规的EIDIG框架。我们的技术结合了快速生成一组多样化的判别种子的全局生成阶段和在模型输出梯度的指导下在这些种子周围生成尽可能多的个体判别实例的局部生成阶段。在每个阶段，充分利用连续迭代的先验信息，加快迭代优化的收敛速度或降低梯度计算的频率。实验结果表明，与现有方法相比，EIDIG方法产生的个体歧视实例平均增加了19.11%，加速速度提高了121.49%，在再训练后的精度损失有限的情况下，减少了80.03%的个体歧视。
Deep learning (DL) systems are increasingly deployed for autonomous decision-making in a wide range of applications. Apart from the robustness and safety, fairness is also an important property that a well-designed DL system should have. To evaluate and improve individual fairness of a model, systematic test case generation for identifying individual discriminatory instances in the input space is essential. In this paper, we propose a framework EIDIG for efficiently discovering individual fairness violation. Our technique combines a global generation phase for rapidly generating a set of diverse discriminatory seeds with a local generation phase for generating as many individual discriminatory instances as possible around these seeds under the guidance of the gradient of the model output. In each phase, prior information at successive iterations is fully exploited to accelerate convergence of iterative optimization or reduce frequency of gradient calculation. Our experimental results show that, on average, our approach EIDIG generates 19.11% more individual discriminatory instances with a speedup of 121.49% when compared with the state-of-the-art method and mitigates individual discrimination by 80.03% with a limited accuracy loss after retraining.