论文标题
塔拉：培训和代表性改变，以实现AI公平和领域的概括
TARA: Training and Representation Alteration for AI Fairness and Domain Generalization
论文作者
论文摘要
我们提出了一种针对受保护或敏感因素实施AI公平性的新方法。该方法使用双重策略执行训练和表示改变（TARA）来缓解AI偏见的显着原因： b）通过使用生成模型，通过域名通过域的适应性和潜在的空间操纵来允许对与代表性不足的人群相关的敏感因素的生成模型，通过智能增强来解决偏见的数据失衡，以解决偏见的数据失衡。在测试图像分析的方法时，实验表明，塔拉（Tara）显着或完全完全clot以基准模型，而在表现优于具有相同信息量的竞争性偏差方法，例如（总体准确性，％精度差距为％，准确性gap）=（78.8，0.5，0.5，0.5）与基线方法相比（71.8，10.5），for eyepac，（79。），以及（79. 7.（79.），以及（73. 73.），（73. 73.），以及（73. 73.），以及（73.），以及（73. 73.）。 21.7）对于Celeba。此外，我们认识到用于评估偏见性能的当前指标中的某些局限性，我们提出了新型的结合性偏见指标。我们的实验还证明了这些新型指标在评估所提出方法的帕累托效率方面的能力。
We propose a novel method for enforcing AI fairness with respect to protected or sensitive factors. This method uses a dual strategy performing training and representation alteration (TARA) for the mitigation of prominent causes of AI bias by including: a) the use of representation learning alteration via adversarial independence to suppress the bias-inducing dependence of the data representation from protected factors; and b) training set alteration via intelligent augmentation to address bias-causing data imbalance, by using generative models that allow the fine control of sensitive factors related to underrepresented populations via domain adaptation and latent space manipulation. When testing our methods on image analytics, experiments demonstrate that TARA significantly or fully debiases baseline models while outperforming competing debiasing methods that have the same amount of information, e.g., with (% overall accuracy, % accuracy gap) = (78.8, 0.5) vs. the baseline method's score of (71.8, 10.5) for EyePACS, and (73.7, 11.8) vs. (69.1, 21.7) for CelebA. Furthermore, recognizing certain limitations in current metrics used for assessing debiasing performance, we propose novel conjunctive debiasing metrics. Our experiments also demonstrate the ability of these novel metrics in assessing the Pareto efficiency of the proposed methods.
