Timezone: »

BARACK: Partially Supervised Group Robustness With Guarantees
Nimit Sohoni · Maziar Sanjabi · Nicolas Ballas · Aditya Grover · Shaoliang Nie · Hamed Firooz · Christopher Re
Event URL: https://openreview.net/forum?id=Rn9POk3wOiV »

While neural networks have shown remarkable success on classification tasks in terms of average-case performance, they often fail to perform well on certain groups of the data, for instance when spurious correlations are present. Unfortunately, group information may be expensive to obtain; thus, recent works in robustness and fairness have proposed ways to improve worst-group performance even when group labels are unavailable. However, these methods generally underperform methods that utilize group information at training time. In this work, we assume access to a small number of group labels alongside a larger dataset without group labels. We propose BARACK, a simple two-step framework to utilize this partial group information to improve worst-group performance: train a model to predict the missing group labels for the training data, and then use these predicted group labels in a robust optimization objective. Theoretically, we provide generalization bounds for our approach in terms of the worst-group performance, which scale with respect to both the total number of training points and the number of training points with group labels. Empirically, across four spurious correlation and robustness benchmark tasks, our method outperforms the baselines that do not use group information, even when only 1-33% of points have group labels.

Author Information

Nimit Sohoni (Stanford University)
Maziar Sanjabi (Meta AI)
Nicolas Ballas (Université de Montréal)
Aditya Grover (UCLA)
Shaoliang Nie (Facebook)
Hamed Firooz (Facebook)
Christopher Re (Stanford University)

More from the Same Authors