Dynamic fairness - Breaking vicious cycles in automatic decision making
This addresses fairness concerns in sensitive decision-making processes, highlighting a critical limitation in existing fairness definitions and proposing a solution for policymakers and practitioners, though it is incremental as it builds on prior research.
The paper tackles the problem of long-term unfairness in automated decision-making by showing that even accurate classifiers adhering to most fairness definitions can perpetuate inequalities over time due to vicious cycles, and finds that only demographic parity avoids this, leading to perfectly accurate and fair classification in the long term.
In recent years, machine learning techniques have been increasingly applied in sensitive decision making processes, raising fairness concerns. Past research has shown that machine learning may reproduce and even exacerbate human bias due to biased training data or flawed model assumptions, and thus may lead to discriminatory actions. To counteract such biased models, researchers have proposed multiple mathematical definitions of fairness according to which classifiers can be optimized. However, it has also been shown that the outcomes generated by some fairness notions may be unsatisfactory. In this contribution, we add to this research by considering decision making processes in time. We establish a theoretic model in which even perfectly accurate classifiers which adhere to almost all common fairness definitions lead to stable long-term inequalities due to vicious cycles. Only demographic parity, which enforces equal rates of positive decisions across groups, avoids these effects and establishes a virtuous cycle, which leads to perfectly accurate and fair classification in the long term.