Simplifying Adversarially
Simplifying Adversarially Robust PAC Learning with Tolerance
Simplifying Adversarially Robust PAC Learning with Tolerance
arXiv:2502.07232v1 Announce Type: new
Abstract: Adversarially robust PAC learning has proved to be challenging, with the currently best known learners [Montasser et al., 2021a] relying on improper methods based on intricate compression schemes, resulting in sample complexity exponential in the VC-d…