Beyond Static Classification: Long-term Fairness for Minority Groups via Performative Prediction and Distributionally Robust Optimization

Abstract

In recent years machine learning (ML) models have begun to be deployed at enormous scales, but too often without adequate concern for whether or not an ML model will make fair decisions. Fairness in ML is a burgeoning research area, but work to define formal fairness criteria has some serious limitations. This thesis aims to combine and explore two recent areas of research in ML – distributionally robust optimization (DRO) and performative prediction – in an attempt to resolve some of these limitations. Performative prediction is a recent framework developed to understand the effects of when deploying model influences the distribution on which it is making predictions, an important concern for fairness. Research on performative prediction has thus far only examined risk minimization, however, which has the potential to result in discriminatory models when working with heterogeneous data composed of majority and minority subgroups. We examine performative prediction with a distributionally robust objective and prove an analogous convergence result for what we call repeated distributionally robust optimization (RDRO). We then verify our results empirically and develop experiments to demonstrate the impact of using RDRO on learning fair ML models.