Enhancing Adversarial Robustness via Test-time Transformation Ensembling
AuthorsPerez, Juan C.
Thabet, Ali Kassem
KAUST DepartmentComputer, Electrical and Mathematical Science and Engineering (CEMSE) Division
Electrical and Computer Engineering
Electrical and Computer Engineering Program
GCR - Award Administration
Office of Competitive Research Funds
VCC Analytics Research Group
Visual Computing Center (VCC)
Permanent link to this recordhttp://hdl.handle.net/10754/670356
MetadataShow full item record
AbstractDeep learning models are prone to being fooled by imperceptible perturbations known as adversarial attacks. In this work, we study how equipping models with Test-time Transformation Ensembling (TTE) can work as a reliable defense against such attacks. While transforming the input data, both at train and test times, is known to enhance model performance, its effects on adversarial robustness have not been studied. Here, we present a comprehensive empirical study of the impact of TTE, in the form of widely-used image transforms, on adversarial robustness. We show that TTE consistently improves model robustness against a variety of powerful attacks without any need for re-training, and that this improvement comes at virtually no trade-off with accuracy on clean samples. Finally, we show that the benefits of TTE transfer even to the certified robustness domain, in which TTE provides sizable and consistent improvements.
CitationPerez, J. C., Alfarra, M., Jeanneret, G., Rueda, L., Thabet, A., Ghanem, B., & Arbelaez, P. (2021). Enhancing Adversarial Robustness via Test-time Transformation Ensembling. 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW). doi:10.1109/iccvw54120.2021.00015
SponsorsThis work was partially supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research.
Conference/Event name18th IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021