Offline Contextual Bandits with High Probability Fairness Guarantees

We present RobinHood, an offline contextual bandit algorithm designed to satisfy a broad family of fairness constraints. Our algorithm accepts multiple fairness definitions and allows users to construct their own unique fairness definitions for the problem at hand. We provide a theoretical analysis of RobinHood, which includes a proof that it will not return an unfair solution with probability greater than a user-specified threshold. We validate our algorithm on three applications: a user study with an automated tutoring system, a loan approval setting using the Statlog German credit data set, and a criminal recidivism problem using data released by ProPublica. To demonstrate the versatility of our approach, we use multiple well-known and custom definitions of fairness. In each setting, our algorithm is able to produce fair policies that achieve performance competitive with other offline and online contextual bandit algorithms.