Skip to yearly menu bar Skip to main content


Exploring the Landscape of Spatial Robustness

Logan Engstrom · Brandon Tran · Dimitris Tsipras · Ludwig Schmidt · Aleksander Madry

Pacific Ballroom #142

Keywords: [ Adversarial Examples ]

Abstract: The study of adversarial robustness has so far largely focused on perturbations bound in $\ell_p$-norms. However, state-of-the-art models turn out to be also vulnerable to other, more natural classes of perturbations such as translations and rotations. In this work, we thoroughly investigate the vulnerability of neural network--based classifiers to rotations and translations. While data augmentation offers relatively small robustness, we use ideas from robust optimization and test-time input aggregation to significantly improve robustness. Finally we find that, in contrast to the $\ell_p$-norm case, first-order methods cannot reliably find worst-case perturbations. This highlights spatial robustness as a fundamentally different setting requiring additional study.

Live content is unavailable. Log in and register to view live content