Making predictions that are fair with regard to protected group membership (race, gender, age, etc.) has become an important requirement for classification algorithms. Existing techniques derive a fair model from sampled labeled data relying on the assumption that training and testing data are identically and independently drawn (iid) from the same distribution.In practice, distribution shift can and does occur between training and testing datasets as the characteristics of individuals interacting with the machine learning system -- and which individuals interact with the system -- change. We investigate fairness under covariate shift, a relaxation of the iid assumption in which the inputs or covariates change while the conditional label distribution remains the same. We seek fair decisions under these assumptions on target data with unknown labels.We propose an approach that obtains the predictor that is robust to the worst-case in terms of target performance while satisfying target fairness requirements and matching statistical properties of the source data. We demonstrate the benefits of our approach on benchmark prediction tasks.

翻译：对受保护群体成员(种族、性别、年龄等)作出公平的预测已成为分类算法的一个重要要求。现有技术从抽样标签数据中得出一个公平的模型,依据的假设是,培训和测试数据是相同和独立地从同一分布中提取的(二d)。 在实践上,分布转移可以而且确实发生在培训和测试数据集之间,作为与机器学习系统互动的个人的特征 -- -- 以及哪些个人与系统互动的特征 -- -- 变化。我们在共变转换中调查公平性,放松在有条件标签分配保持不变的情况下投入或共变的假设。我们寻求在这些假设下对有未知标签的目标数据作出公平决定。我们提出一种办法,在满足目标公平性要求和匹配源数据统计特性的同时,获得在目标性业绩方面最坏情况的可靠预测者。我们展示了我们在基准预测任务方面的做法的好处。