[edit]
Adversarially Robust PAC Learnability of Real-Valued Functions
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:1172-1199, 2023.
Abstract
We study robustness to test-time adversarial attacks in the regression setting with ℓp losses and arbitrary perturbation sets. We address the question of which function classes are PAC learnable in this setting. We show that classes of finite fat-shattering dimension are learnable in both the realizable and agnostic settings. Moreover, for convex function classes, they are even properly learnable. In contrast, some non-convex function classes provably require improper learning algorithms. Our main technique is based on a construction of an adversarially robust sample compression scheme of a size determined by the fat-shattering dimension. Along the way, we introduce a novel agnostic sample compression scheme for real-valued functions, which may be of independent interest.