Toward a better trade-off between performance and fairness with kernel-based distribution matching
arXiv:1910.11779
Abstract
As recent literature has demonstrated how classifiers often carry unintended biases toward some subgroups, deploying machine learned models to users demands careful consideration of the social consequences. How should we address this problem in a real-world system? How should we balance core performance and fairness metrics? In this paper, we introduce a MinDiff framework for regularizing classifiers toward different fairness metrics and analyze a technique with kernel-based statistical dependency tests. We run a thorough study on an academic dataset to compare the Pareto frontier achieved by different regularization approaches, and apply our kernel-based method to two large-scale industrial systems demonstrating real-world improvements.
References in corpus (9)
- Learning Transferable Features with Deep Adaptation Networks
- Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
- Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
- A Kernel Method for the Two-Sample Problem
- Learning Adversarially Fair and Transferable Representations
- Satisfying Real-world Goals with Dataset Constraints
- Does mitigating ML's impact disparity require treatment disparity?
- The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the xAUC Metric
- Transfer of Machine Learning Fairness across Domains