FORML: Learning to Reweight Data for Fairness
It addresses fairness violations in imbalanced data for users of ML models, offering an incremental improvement by integrating fairness into training without architectural changes.
The paper tackles the problem of machine learning models neglecting fairness and robustness by introducing FORML, a training algorithm that jointly learns sample weights and model parameters to balance fairness with accuracy, achieving improvements in fairness criteria on image classification tasks without sacrificing accuracy.
Machine learning models are trained to minimize the mean loss for a single metric, and thus typically do not consider fairness and robustness. Neglecting such metrics in training can make these models prone to fairness violations when training data are imbalanced or test distributions differ. This work introduces Fairness Optimized Reweighting via Meta-Learning (FORML), a training algorithm that balances fairness and robustness with accuracy by jointly learning training sample weights and neural network parameters. The approach increases model fairness by learning to balance the contributions from both over- and under-represented sub-groups through dynamic reweighting of the data learned from a user-specified held-out set representative of the distribution under which fairness is desired. FORML improves equality of opportunity fairness criteria on image classification tasks, reduces bias of corrupted labels, and facilitates building more fair datasets via data condensation. These improvements are achieved without pre-processing data or post-processing model outputs, without learning an additional weighting function, without changing model architecture, and while maintaining accuracy on the original predictive metric.