Fairness with Continuous Optimal Transport
This work provides an incremental improvement for machine learning practitioners dealing with fairness concerns, particularly when data availability is limited.
This paper addresses fairness issues in machine learning by introducing a stochastic-gradient method based on continuous optimal transport (OT), extending previous work that was limited to discrete OT. The method demonstrates superior performance compared to discrete OT when data is scarce and comparable performance otherwise. Both continuous and discrete OT methods are shown to adapt model parameters to varying levels of unfairness.
Whilst optimal transport (OT) is increasingly being recognized as a powerful and flexible approach for dealing with fairness issues, current OT fairness methods are confined to the use of discrete OT. In this paper, we leverage recent advances from the OT literature to introduce a stochastic-gradient fairness method based on a dual formulation of continuous OT. We show that this method gives superior performance to discrete OT methods when little data is available to solve the OT problem, and similar performance otherwise. We also show that both continuous and discrete OT methods are able to continually adjust the model parameters to adapt to different levels of unfairness that might occur in real-world applications of ML systems.