Timezone: »

Subgroup Robustness Grows On Trees: An Empirical Baseline Investigation
Josh Gardner · Zoran Popovic · Ludwig Schmidt

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #623

Researchers have proposed many methods for fair and robust machine learning, but thorough empirical evaluation of their subgroup robustness is lacking. In this work, we address this gap in the context of tabular data, where sensitive subgroups are clearly-defined, real-world fairness problems abound, and prior works often fail to compare to state-of-the-art tree-based models. We conduct an empirical comparison of several previously-proposed methods for fair and robust learning alongside state-of-the-art tree-based methods and other baselines. Via experiments with more than 340,000 model configurations on eight datasets, we show that tree-based methods have strong subgroup robustness, even when compared to robustness- and fairness-enhancing methods. Moreover, the best tree-based models tend to show good performance over a range of metrics, while robust or group-fair models can show brittleness, with significant performance differences across different metrics for a fixed model. We also demonstrate that tree-based models show less sensitivity to hyperparameter configurations, and are less costly to train. Our work suggests that tree-based ensemble models make an effective baseline for tabular data, and are a sensible default when subgroup robustness is desired.

Author Information

Josh Gardner (University of Washington)
Zoran Popovic (University of Washington)
Ludwig Schmidt (University of Washington)

More from the Same Authors