Metric-Free Individual Fairness with Cooperative Contextual Bandits
This addresses bias in machine learning for applications like automated decision-making, offering a novel method to enforce individual fairness without predefined similarity metrics, though it is incremental in combining fairness with bandit algorithms.
The paper tackles the problem of bias in automated decision-making by proposing a metric-free individual fairness approach using cooperative contextual bandits, which treats fairness as a reward to mitigate bias without requiring differentiable metrics, and shows effectiveness on real-world datasets in achieving both individual and group fairness.
Data mining algorithms are increasingly used in automated decision making across all walks of daily life. Unfortunately, as reported in several studies these algorithms inject bias from data and environment leading to inequitable and unfair solutions. To mitigate bias in machine learning, different formalizations of fairness have been proposed that can be categorized into group fairness and individual fairness. Group fairness requires that different groups should be treated similarly which might be unfair to some individuals within a group. On the other hand, individual fairness requires that similar individuals be treated similarly. However, individual fairness remains understudied due to its reliance on problem-specific similarity metrics. We propose a metric-free individual fairness and a cooperative contextual bandits (CCB) algorithm. The CCB algorithm utilizes fairness as a reward and attempts to maximize it. The advantage of treating fairness as a reward is that the fairness criterion does not need to be differentiable. The proposed algorithm is tested on multiple real-world benchmark datasets. The results show the effectiveness of the proposed algorithm at mitigating bias and at achieving both individual and group fairness.