Robust Fairness-aware Learning Under Sample Selection Bias
This work addresses fairness in machine learning for scenarios with biased training data, which is an incremental improvement over existing bias correction methods.
The paper tackles the problem of achieving fairness in classification models under sample selection bias, where training and test data distributions differ, by proposing a framework that combines reweighing for bias correction and minimax robust estimation to ensure fairness under worst-case scenarios, with experimental results on two real-world datasets showing effectiveness in utility and fairness metrics.
The underlying assumption of many machine learning algorithms is that the training data and test data are drawn from the same distributions. However, the assumption is often violated in real world due to the sample selection bias between the training and test data. Previous research works focus on reweighing biased training data to match the test data and then building classification models on the reweighed training data. However, how to achieve fairness in the built classification models is under-explored. In this paper, we propose a framework for robust and fair learning under sample selection bias. Our framework adopts the reweighing estimation approach for bias correction and the minimax robust estimation approach for achieving robustness on prediction accuracy. Moreover, during the minimax optimization, the fairness is achieved under the worst case, which guarantees the model's fairness on test data. We further develop two algorithms to handle sample selection bias when test data is both available and unavailable. We conduct experiments on two real-world datasets and the experimental results demonstrate its effectiveness in terms of both utility and fairness metrics.