ICML 2020poster40 citations
Efficiently Learning Adversarially Robust Halfspaces with Noise
Omar Montasser, Surbhi Goel, Ilias Diakonikolas, Nathan Srebro
Abstract
We study the problem of learning adversarially robust halfspaces in the distribution-independent setting. In the realizable setting, we provide necessary and sufficient conditions on the adversarial perturbation sets under which halfspaces are efficiently robustly learnable. In the presence of random label noise, we give a simple computationally efficient algorithm for this problem with respect to any $\ell_p$-perturbation.
BibTeX
@InProceedings{pmlr-v119-montasser20a,
title = {Efficiently Learning Adversarially Robust Halfspaces with Noise},
author = {Montasser, Omar and Goel, Surbhi and Diakonikolas, Ilias and Srebro, Nathan},
booktitle = {Proceedings of the 37th International Conference on Machine Learning},
pages = {7010--7021},
year = {2020},
editor = {III, Hal Daumé and Singh, Aarti},
volume = {119},
series = {Proceedings of Machine Learning Research},
month = {13--18 Jul},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v119/montasser20a/montasser20a.pdf},
url = {https://proceedings.mlr.press/v119/montasser20a.html},
abstract = {We study the problem of learning adversarially robust halfspaces in the distribution-independent setting. In the realizable setting, we provide necessary and sufficient conditions on the adversarial perturbation sets under which halfspaces are efficiently robustly learnable. In the presence of random label noise, we give a simple computationally efficient algorithm for this problem with respect to any $\ell_p$-perturbation.}
}