论文标题
更公平的图像分类的可区分距离近似
A Differentiable Distance Approximation for Fairer Image Classification
论文作者
论文摘要
训练有素的AI模型可能会严重偏见。当偏见涉及法律或道德保护的属性，例如种族背景，年龄或性别时，这可能尤其有问题。现有的解决问题的解决方案是以额外的计算为代价，不稳定的对抗优化或对特征空间结构造成的损失，这些损失与公平措施脱节，并且只能松散地推广到公平。在这项工作中，我们提出了人口统计学方差的可区分近似值，人口统计学的差异可以用来测量AI模型中的偏见或不公平性。我们的近似值可以与常规训练目标一起进行优化，从而消除了在训练过程中对任何额外模型的需求，并直接改善了正规模型的公平性。我们证明我们的方法可以提高AI模型在各种任务和数据集方案中的公平性，同时仍保持高水平的分类精度。代码可从https://bitbucket.org/nelliottrosa/base_fairness获得。
Naively trained AI models can be heavily biased. This can be particularly problematic when the biases involve legally or morally protected attributes such as ethnic background, age or gender. Existing solutions to this problem come at the cost of extra computation, unstable adversarial optimisation or have losses on the feature space structure that are disconnected from fairness measures and only loosely generalise to fairness. In this work we propose a differentiable approximation of the variance of demographics, a metric that can be used to measure the bias, or unfairness, in an AI model. Our approximation can be optimised alongside the regular training objective which eliminates the need for any extra models during training and directly improves the fairness of the regularised models. We demonstrate that our approach improves the fairness of AI models in varied task and dataset scenarios, whilst still maintaining a high level of classification accuracy. Code is available at https://bitbucket.org/nelliottrosa/base_fairness.
