The Frontiers of Fairness in Machine Learning
Alexandra Chouldechova Heinz College, Carnegie Mellon University. Aaron Roth Department of Computer and Information Science, University of Pennsylvania.
Abstract
The last few years have seen an explosion of academic and popular interest in algorithmic fairness. Despite this interest and the volume and velocity of work that has been produced recently, the fundamental science of fairness in machine learning is still in a nascent state. In March 2018, we convened a group of experts as part of a CCC visioning workshop to assess the state of the field, and distill the most promising research directions going forward. This report summarizes the findings of that workshop. Along the way, it surveys recent theoretical work in the field and points towards promising directions for research.
中文速览
机器学习算法在贷款审批、假释决策等高风险场景中的广泛应用，引发了人们对算法歧视（algorithmic bias）的强烈担忧。2018年，约50位来自学界、业界和政府的专家召开研讨会，系统梳理了算法公平性（algorithmic fairness）领域的现有认知与核心难题：训练数据本身携带历史偏见、最小化平均误差天然偏向多数群体、探索性决策成本由弱势群体不成比例地承担，这些都是算法不公平的根源。在公平性定义上，统计公平（statistical fairness）易于验证但保障粒度粗糙，个体公平（individual fairness）语义更强却依赖难以获得的先验假设，两类方法各有短板且往往无法同时满足。研讨会进一步指出，动态环境下公平性如何随时间演化、多组件系统中公平性能否稳健合成、激励机制与人的行为反馈如何影响公平目标，都是亟待突破的前沿方向。这份报告的价值在于，它以理论视角为算法公平性研究划定了科学边界，为后续工作提供了系统性的问题图谱。
原文 arXiv:1810.08810;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1810.08810v1