Beware the Black-Box: On the Robustness of Recent Defenses to Adversarial Examples |
| |
Authors: | Kaleel Mahmood Deniz Gurevin Marten van Dijk Phuoung Ha Nguyen |
| |
Institution: | 1.Department of Computer Science and Engineering, University of Connecticut, Storrs, CT 06269, USA;2.Department of Electrical and Computer Engineering, University of Connecticut, Storrs, CT 06269, USA;3.CWI, 1098 XG Amsterdam, The Netherlands;4.eBay, San Jose, CA 95125, USA; |
| |
Abstract: | Many defenses have recently been proposed at venues like NIPS, ICML, ICLR and CVPR. These defenses are mainly focused on mitigating white-box attacks. They do not properly examine black-box attacks. In this paper, we expand upon the analyses of these defenses to include adaptive black-box adversaries. Our evaluation is done on nine defenses including Barrage of Random Transforms, ComDefend, Ensemble Diversity, Feature Distillation, The Odds are Odd, Error Correcting Codes, Distribution Classifier Defense, K-Winner Take All and Buffer Zones. Our investigation is done using two black-box adversarial models and six widely studied adversarial attacks for CIFAR-10 and Fashion-MNIST datasets. Our analyses show most recent defenses (7 out of 9) provide only marginal improvements in security (<25%), as compared to undefended networks. For every defense, we also show the relationship between the amount of data the adversary has at their disposal, and the effectiveness of adaptive black-box attacks. Overall, our results paint a clear picture: defenses need both thorough white-box and black-box analyses to be considered secure. We provide this large scale study and analyses to motivate the field to move towards the development of more robust black-box defenses. |
| |
Keywords: | adversarial machine learning black-box attacks security |
|
|