Learning from Discriminatory Training Data
This addresses the issue of algorithmic fairness for protected groups, particularly in legal and business contexts, though it appears incremental as it builds on existing fair learning methods.
The paper tackles the problem of supervised learning systems unintentionally learning discrimination from biased training data, proposing a method that provably minimizes model error on fair test datasets while training on data poisoned with direct additive discrimination.
Supervised learning systems are trained using historical data and, if the data was tainted by discrimination, they may unintentionally learn to discriminate against protected groups. We propose that fair learning methods, despite training on potentially discriminatory datasets, shall perform well on fair test datasets. Such dataset shifts crystallize application scenarios for specific fair learning methods. For instance, the removal of direct discrimination can be represented as a particular dataset shift problem. For this scenario, we propose a learning method that provably minimizes model error on fair datasets, while blindly training on datasets poisoned with direct additive discrimination. The method is compatible with existing legal systems and provides a solution to the widely discussed issue of protected groups' intersectionality by striking a balance between the protected groups. Technically, the method applies probabilistic interventions, has causal and counterfactual formulations, and is computationally lightweight - it can be used with any supervised learning model to prevent direct and indirect discrimination via proxies while maximizing model accuracy for business necessity.