Mind the Graph When Balancing Data for Fairness or Robustness

Jessica Schrouff, Alexis Bellot, Amal Rannen-Triki, Alan Malek, Isabela Albuquerque, Arthur Gretton, Alexander Nicholas D'Amour, Silvia Chiappa

Keywords: Fairness, Robustness, Data balancing, Causality, Fine-tuning

TL;DR: We use a causal framework to investigate the failure modes of data balancing for fairness or robustness, its conditions for success, and its interaction with other mitigation strategies.

Abstract: Failures of fairness or robustness in machine learning predictive settings can be due to undesired dependencies between covariates, outcomes and auxiliary factors of variation. A common strategy to mitigate these failures is data balancing, which attempts to remove those undesired dependencies. In this work, we define conditions on the training distribution for data balancing to lead to fair or robust models. Our results display that in many cases, the balanced distribution does not correspond to selectively removing the undesired dependencies in a causal graph of the task, leading to multiple failure modes and even interference with other mitigation techniques such as regularization. Overall, our results highlight the importance of taking the causal graph into account before performing data balancing.

Primary Area: Fairness

Submission Number: 3374