Mitigating Spurious Correlation via Distributionally Robust Learning with Hierarchical Ambiguity Sets
This addresses spurious correlation issues for machine learning models, especially in minority groups with limited data, representing an incremental improvement over existing robust learning methods.
The paper tackles the problem of spurious correlations in machine learning, particularly under intra-group distribution shifts in minority groups, by proposing a hierarchical extension of Group DRO that addresses both inter-group and intra-group uncertainties, achieving strong robustness where existing methods fail and superior performance on standard benchmarks.
Conventional supervised learning methods are often vulnerable to spurious correlations, particularly under distribution shifts in test data. To address this issue, several approaches, most notably Group DRO, have been developed. While these methods are highly robust to subpopulation or group shifts, they remain vulnerable to intra-group distributional shifts, which frequently occur in minority groups with limited samples. We propose a hierarchical extension of Group DRO that addresses both inter-group and intra-group uncertainties, providing robustness to distribution shifts at multiple levels. We also introduce new benchmark settings that simulate realistic minority group distribution shifts-an important yet previously underexplored challenge in spurious correlation research. Our method demonstrates strong robustness under these conditions-where existing robust learning methods consistently fail-while also achieving superior performance on standard benchmarks. These results highlight the importance of broadening the ambiguity set to better capture both inter-group and intra-group distributional uncertainties.