Brown University（布朗大学）；Northeastern University（东北大学）；Carnegie Mellon University（卡内基梅隆大学）
Jessica Dai, Sina Fazelpour, Zachary C. Lipton
备注：
Presented at AIES 2021; previously at the NeurIPS 2020 Workshop on Consequential Decision Making in Dynamic Environments and the NeurIPS 2020 Workshop on ML for Economic Policy. Minor correction uploaded Sept. 2022
Typically, fair machine learning research focuses on a single decisionmaker and assumes that the underlying population is stationary. However, many of the critical domains motivating this work are characterized by competitive marketplaces with many decisionmakers. Realistically, we might expect only a subset of them to adopt any non-compulsory fairness-conscious policy, a situation that political philosophers call partial compliance. This possibility raises important questions: how does the strategic behavior of decision subjects in partial compliance settings affect the allocation outcomes? If k% of employers were to voluntarily adopt a fairness-promoting intervention, should we expect k% progress (in aggregate) towards the benefits of universal adoption, or will the dynamics of partial compliance wash out the hoped-for benefits? How might adopting a global (versus local) perspective impact the conclusions of an auditor? In this paper, we propose a simple model of an employment market, leveraging simulation as a tool to explore the impact of both interaction effects and incentive effects on outcomes and auditing metrics. Our key findings are that at equilibrium: (1) partial compliance (k% of employers) can result in far less than proportional (k%) progress towards the full compliance outcomes; (2) the gap is more severe when fair employers match global (vs local) statistics; (3) choices of local vs global statistics can paint dramatically different pictures of the performance vis-a-vis fairness desiderata of compliant versus non-compliant employers; and (4) partial compliance to local parity measures can induce extreme segregation.
Co-designing for Compliance: Multi-party Computation Protocols for Post-Market Fairness Monitoring in Algorithmic Hiring
Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））；Max Planck Institute for Security and Privacy（马克斯·普朗克安全与隐私研究所）；Leiden University（莱顿大学）；Universitat Pompeu Fabra（庞培法布拉大学）；InfoJobs；Ruhr University Bochum（波鸿鲁尔大学）；Radboud University（拉德堡德大学）
Beyond Static Knowledge Messengers: Towards Adaptive, Fair, and Scalable Federated Learning for Medical AI
Department of Computer Science and Software Engineering, Auburn University（计算机科学与软件工程系，阿伯丁大学）；Department of Information and Communication Technology, Bangladesh University of Professionals（信息与通信技术系，孟加拉国专业大学）；Mymensingh Medical College and Hospital（迈明辛医疗学院和医院）
Data Bias Mitigation under Coverage Constraints & The Price of Fairness
Khoury College of Computer Sciences, Northeastern University（东北大学库里计算机科学学院）；Cheriton School of Computer Science, University of Waterloo（滑铁卢大学切里顿计算机科学学院）
flowengineR: A Modular and Extensible Framework for Fair and Reproducible Workflow Design in R
Institute for Mathematics, School of Mathematics and Sciences, Carl von Ossietzky University Oldenburg（奥尔登堡卡尔·冯·奥西茨基大学数学与科学学院数学研究所）；HASPA Finanzholding（HASPA金融控股公司）
Mathematical Notions vs. Human Perception of Fairness: A Descriptive Approach to Fairness for Machine Learning
最近，机器学习的公平性受到了相当大的关注。已经提出了各种公平的数学公式，并且已经表明不可能同时满足所有这些公式。到目前为止，文献已经通过量化不同公平公式之间的权衡来处理这些不可能的结果。我们的工作从不同的角度看待这个问题。我们不是要求所有公平的概念（部分）同时成立，而是考虑到决策模型将被部署的社会领域，其中哪一个是最合适的。我们采取描述性的方法，并着手确定公平的概念，最好的捕捉\emph{lay people's perception of fairness}。我们进行了适应性实验，旨在通过少量的测试来确定与每个参与者的选择最兼容的公平概念。也许令人惊讶的是，我们发现公平的最简单的数学定义-即人口均等-在两个不同的应用场景中最接近人们对公平的想法。即使我们明确地告诉参与者关于公平的另一种更复杂的定义，并且我们减轻了他们评估这些概念的认知负担，这个结论仍然完好无损。我们的研究结果对公平ML文献和关于形式化算法公平性的讨论具有重要意义。
