Fairness in Reinforcement Learning

Shahin Jabbari, Matthew Joseph, Michael Kearns +2

We initiate the study of fairness in reinforcement learning, where the actions of a learning algorithm may affect its environment and future rewards. Our fairness constraint requir…