Agnostic Multi-Robust Learning Using ERM
This work addresses the challenge of asymmetry in robust learning for machine learning practitioners, though it is incremental as it builds on prior algorithms.
The paper tackles the problem of robust learning when no classifier achieves zero robust error, extending previous work to the non-robustly-realizable case and introducing a multi-group setting for robust loss across subgroups.
A fundamental problem in robust learning is asymmetry: a learner needs to correctly classify every one of exponentially-many perturbations that an adversary might make to a test-time natural example. In contrast, the attacker only needs to find one successful perturbation. Xiang et al.[2022] proposed an algorithm that in the context of patch attacks for image classification, reduces the effective number of perturbations from an exponential to a polynomial number of perturbations and learns using an ERM oracle. However, to achieve its guarantee, their algorithm requires the natural examples to be robustly realizable. This prompts the natural question; can we extend their approach to the non-robustly-realizable case where there is no classifier with zero robust error? Our first contribution is to answer this question affirmatively by reducing this problem to a setting in which an algorithm proposed by Feige et al.[2015] can be applied, and in the process extend their guarantees. Next, we extend our results to a multi-group setting and introduce a novel agnostic multi-robust learning problem where the goal is to learn a predictor that achieves low robust loss on a (potentially) rich collection of subgroups.