PC-Fairness: A Unified Framework for Measuring Causality-based Fairness
This work addresses the identifiability problem in causality-based fairness for machine learning practitioners, offering a general solution but is incremental as it builds on existing notions.
The paper tackles the challenge of measuring causality-based fairness notions from observational data by proposing a unified framework called path-specific counterfactual fairness (PC fairness) and a method to bound fairness under unidentifiable situations, with experiments on synthetic and real-world datasets demonstrating correctness and effectiveness.
A recent trend of fair machine learning is to define fairness as causality-based notions which concern the causal connection between protected attributes and decisions. However, one common challenge of all causality-based fairness notions is identifiability, i.e., whether they can be uniquely measured from observational data, which is a critical barrier to applying these notions to real-world situations. In this paper, we develop a framework for measuring different causality-based fairness. We propose a unified definition that covers most of previous causality-based fairness notions, namely the path-specific counterfactual fairness (PC fairness). Based on that, we propose a general method in the form of a constrained optimization problem for bounding the path-specific counterfactual fairness under all unidentifiable situations. Experiments on synthetic and real-world datasets show the correctness and effectiveness of our method.