arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.06004cs.LGstat.ML
收藏

公平感知的PAC从受损数据中学习

Fairness-Aware PAC Learning from Corrupted Data

ETH AI Center(苏黎世联邦理工学院人工智能中心);Institute of Science and Technology Austria (ISTA)(奥地利科学技术研究所(ISTA))

Nikola Konstantinov, Christoph H. Lampert

摘要:

解决关于机器学习模型的公平性问题是在现实世界的自动化系统中长期采用它们的关键一步。虽然已经开发了许多方法来从数据中训练公平模型,但对这些方法对数据损坏的鲁棒性知之甚少。在这项工作中,我们考虑在最坏情况下的数据操作的公平意识的学习。我们表明,在某些情况下,对手可以迫使任何学习者返回一个过度偏见的分类器,无论样本大小,并与或不降低准确性,以及过度偏见的强度增加与数据中代表性不足的保护组的学习问题。我们还证明了我们的硬度结果是紧的常数因子。为此,我们研究了两个自然的学习算法,优化的准确性和公平性,并表明这些算法享有保证,是为了最佳的腐败率和保护组的频率在大数据的限制。

英文摘要:

Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. While many approaches have been developed for training fair models from data, little is known about the robustness of these methods to data corruption. In this work we consider fairness-aware learning under worst-case data manipulations. We show that an adversary can in some situations force any learner to return an overly biased classifier, regardless of the sample size and with or without degrading accuracy, and that the strength of the excess bias increases for learning problems with underrepresented protected groups in the data. We also prove that our hardness results are tight up to constant factors. To this end, we study two natural learning algorithms that optimize for both accuracy and fairness and show that these algorithms enjoy guarantees that are order-optimal in terms of the corruption ratio and the protected groups frequencies in the large data limit.

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。

↑