Confidence-Aware Training of Smoothed Classifiers for Certified Robustness
Jongheon Jeong, Seojin Kim, Jinwoo Shin
Abstract
Any classifier can be "smoothed out" under Gaussian noise to build a new classifier that is provably robust to l2-adversarial perturbations, viz., by averaging its predictions over the noise via randomized smoothing. Under the smoothed classifiers, the fundamental trade-off between accuracy and (adversarial) robustness has been well evidenced in the literature: i.e., increasing the robustness of a classifier for an input can be at the expense of decreased accuracy for some other inputs. In this paper, we propose a simple training method leveraging this trade-off to obtain robust smoothed classifiers, in particular, through a sample-wise control of robustness over the training samples. We make this control feasible by using "accuracy under Gaussian noise" as an easy-to-compute proxy of adversarial robustness for an input. Specifically, we differentiate the training objective depending on this proxy to filter out samples that are unlikely to benefit from the worst-case (adversarial) objective. Our experiments show that the proposed method, despite its simplicity, consistently exhibits improved certified robustness upon state-of-the-art training methods. Somewhat surprisingly, we find these improvements persist even for other notions of robustness, e.g., to various types of common corruptions. Code is available at https://github.com/alinlab/smoothing-catrs.
BibTeX
@article{Jeong_Kim_Shin_2023, title={Confidence-Aware Training of Smoothed Classifiers for Certified Robustness}, volume={37}, url={https://ojs.aaai.org/index.php/AAAI/article/view/25968}, DOI={10.1609/aaai.v37i7.25968}, abstractNote={Any classifier can be "smoothed out" under Gaussian noise to build a new classifier that is provably robust to l2-adversarial perturbations, viz., by averaging its predictions over the noise via randomized smoothing. Under the smoothed classifiers, the fundamental trade-off between accuracy and (adversarial) robustness has been well evidenced in the literature: i.e., increasing the robustness of a classifier for an input can be at the expense of decreased accuracy for some other inputs. In this paper, we propose a simple training method leveraging this trade-off to obtain robust smoothed classifiers, in particular, through a sample-wise control of robustness over the training samples. We make this control feasible by using "accuracy under Gaussian noise" as an easy-to-compute proxy of adversarial robustness for an input. Specifically, we differentiate the training objective depending on this proxy to filter out samples that are unlikely to benefit from the worst-case (adversarial) objective. Our experiments show that the proposed method, despite its simplicity, consistently exhibits improved certified robustness upon state-of-the-art training methods. Somewhat surprisingly, we find these improvements persist even for other notions of robustness, e.g., to various types of common corruptions. Code is available at https://github.com/alinlab/smoothing-catrs.}, number={7}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Jeong, Jongheon and Kim, Seojin and Shin, Jinwoo}, year={2023}, month={Jun.}, pages={8005-8013} }