Incentives Needed for Low-Cost Fair Lateral Data Reuse
Incentives Needed for Low-Cost Fair Lateral Data Reuse
复制标题
DOI：
10.1145/3412815.3416890
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
R. Maio;A. Chaintreau
中科院分区：
文献类型：
--
作者：
R. Maio;A. Chaintreau
算法公平性的一个中心目标是构建具有公平属性的系统，这些系统可以优雅地组成。在数据科学中，实现这一目标的一个主要努力和步骤是开发空气表示，通过施加人口统计保密约束来保证连续组成下的人口统计均等。在这项工作中，我们阐明了人口秘密公平代表的局限性，并提出了一种新的方法来克服这些局限性，将有关各方的激励措施的信息纳入公平干预。具体来说，我们表明，在一个程式化的模型，它是可能的，以获得激励相容的表示，其中理性的各方获得指数更大的效用相对维斯任何人口秘密的表示，并满足人口均等。这些实质性的收益不是从众所周知的公平成本中获得的，而是从我们第一次正式化和量化的人口保密成本中获得的。我们进一步表明，人口秘密表示的顺序组成属性是不强大的聚合。我们的研究结果为公平组合，公平机器学习和算法公平性的研究开辟了几个新的方向。
A central goal of algorithmic fairness is to build systems with fairness properties that compose gracefully. A major effort and step towards this goal in data science has been the development offair representations which guarantee demographic parity under sequential composition by imposing ademographic secrecy constraint. In this work, we elucidate limitations of demographically secret fair representations and propose a fresh approach to potentially overcome them by incorporating information about parties' incentives into fairness interventions. Specifically, we show that in a stylized model, it is possible to relax demographic secrecy to obtainincentive-compatible representations, where rational parties obtain exponentially greater utilities vis-à-vis any demographically secret representation and satisfy demographic parity. These substantial gains are recovered not from the well-knowncost of fairness, but rather from acost of demographic secrecy which we formalize and quantify for the first time. We further show that the sequential composition property of demographically secret representations is not robust to aggregation. Our results open several new directions for research in fair composition, fair machine learning and algorithmic fairness.