cicyt UNIZAR
Full-text links:

Download:

Current browse context:

cs.CV

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: Improving Transferability of Adversarial Examples with Input Diversity

Abstract: Though convolutional neural networks have achieved state-of-the-art performance on various vision tasks, they are extremely vulnerable to adversarial examples, which are obtained by adding human-imperceptible perturbations to the original images. Adversarial examples can thus be used as an useful tool to evaluate and select the most robust models in safety-critical applications. However, most of the existing adversarial attacks only achieve relatively low success rates under the challenging black-box setting, where the attackers have no knowledge of the model structure and parameters. To this end, we propose to improve the transferability of adversarial examples by creating diverse input patterns. Instead of only using the original images to generate adversarial examples, our method applies random transformations to the input images at each iteration. Extensive experiments on ImageNet show that the proposed attack method can generate adversarial examples that transfer much better to different networks than existing baselines. To further improve the transferability, we (1) integrate the recently proposed momentum method into the attack process; and (2) attack an ensemble of networks simultaneously. By evaluating our method against top defense submissions and official baselines from NIPS 2017 adversarial competition, this enhanced attack reaches an average success rate of 73.0%, which outperforms the top 1 attack submission in the NIPS competition by a large margin of 6.6%. We hope that our proposed attack strategy can serve as a benchmark for evaluating the robustness of networks to adversaries and the effectiveness of different defense methods in future. The code is public available at this https URL
Comments: Submitted to ECCV 2018, code available at this https URL
Subjects: Computer Vision and Pattern Recognition (cs.CV); Learning (cs.LG); Machine Learning (stat.ML)
Cite as: arXiv:1803.06978 [cs.CV]
  (or arXiv:1803.06978v1 [cs.CV] for this version)

Submission history

From: Cihang Xie [view email]
[v1] Mon, 19 Mar 2018 15:07:51 GMT (307kb,D)