On Learning Fairness and Accuracy on Multiple Subgroups
This work addresses fairness in machine learning for scenarios with many subgroups, offering a method to balance utility and fairness, though it appears incremental as it builds on existing fair learning frameworks.
The paper tackles the problem of learning fair predictors for multiple subgroups with limited data, proposing a bilevel objective method that reduces prediction disparities while maintaining accuracy, achieving consistently improved fair predictions and comparable accuracy to baselines.
We propose an analysis in fair learning that preserves the utility of the data while reducing prediction disparities under the criteria of group sufficiency. We focus on the scenario where the data contains multiple or even many subgroups, each with limited number of samples. As a result, we present a principled method for learning a fair predictor for all subgroups via formulating it as a bilevel objective. Specifically, the subgroup specific predictors are learned in the lower-level through a small amount of data and the fair predictor. In the upper-level, the fair predictor is updated to be close to all subgroup specific predictors. We further prove that such a bilevel objective can effectively control the group sufficiency and generalization error. We evaluate the proposed framework on real-world datasets. Empirical evidence suggests the consistently improved fair predictions, as well as the comparable accuracy to the baselines.