Contrastive Fairness in Machine Learning
This work addresses fairness in machine learning for real-world decision-making, focusing on contrastive rather than counterfactual questions, which is a novel approach but may be incremental in the broader fairness domain.
The paper tackles the problem of ensuring fairness in algorithmic decision-making by addressing contrastive questions like 'why this but not that?', introducing concepts and mathematical tools using causal inference to handle such scenarios.
Was it fair that Harry was hired but not Barry? Was it fair that Pam was fired instead of Sam? How can one ensure fairness when an intelligent algorithm takes these decisions instead of a human? How can one ensure that the decisions were taken based on merit and not on protected attributes like race or sex? These are the questions that must be answered now that many decisions in real life can be made through machine learning. However research in fairness of algorithms has focused on the counterfactual questions "what if?" or "why?", whereas in real life most subjective questions of consequence are contrastive: "why this but not that?". We introduce concepts and mathematical tools using causal inference to address contrastive fairness in algorithmic decision-making with illustrative examples.