arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.17433cs.LG
收藏

在为公平性或鲁棒性平衡数据时,注意因果图

Mind the Graph When Balancing Data for Fairness or Robustness

Google DeepMind(谷歌DeepMind)

Jessica Schrouff, Alexis Bellot, Amal Rannen-Triki, Alan Malek, Isabela Albuquerque, Arthur Gretton, Alexander D'Amour, Silvia Chiappa

AI总结:

本文研究数据平衡对公平性与鲁棒性的影响,指出若忽视任务因果图,平衡分布可能无法消除不期望依赖,反而引发失效并干扰正则化等方法。

摘要:

机器学习预测环境中公平性或鲁棒性的失效,可能源于协变量、结果与辅助变异因素之间不期望的依赖关系。缓解这些失效的一种常见策略是数据平衡,它试图消除这些不期望的依赖关系。在本工作中,我们定义了训练分布上使数据平衡能够产生公平或鲁棒模型的条件。我们的结果表明,在许多情况下,平衡后的分布并不对应于在任务因果图中有选择地消除不期望的依赖关系,从而导致多种失效模式,甚至干扰其他缓解技术(如正则化)。总体而言,我们的结果强调了在执行数据平衡之前考虑因果图的重要性。

英文摘要:

Failures of fairness or robustness in machine learning predictive settings can be due to undesired dependencies between covariates, outcomes and auxiliary factors of variation. A common strategy to mitigate these failures is data balancing, which attempts to remove those undesired dependencies. In this work, we define conditions on the training distribution for data balancing to lead to fair or robust models. Our results display that, in many cases, the balanced distribution does not correspond to selectively removing the undesired dependencies in a causal graph of the task, leading to multiple failure modes and even interference with other mitigation techniques such as regularization. Overall, our results highlight the importance of taking the causal graph into account before performing data balancing.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑