Enhancing adversarial example transferability with an intermediate level attack
Research output: Contribution to journal › Conference article › Research › peer-review
Neural networks are vulnerable to adversarial examples, malicious inputs crafted to fool trained models. Adversarial examples often exhibit black-box transfer, meaning that adversarial examples for one model can fool another model. However, adversarial examples are typically overfit to exploit the particular architecture and feature representation of a source model, resulting in sub-optimal black-box transfer attacks to other target models. We introduce the Intermediate Level Attack (ILA), which attempts to fine-tune an existing adversarial example for greater black-box transferability by increasing its perturbation on a pre-specified layer of the source model, improving upon state-of-the-art methods. We show that we can select a layer of the source model to perturb without any knowledge of the target models while achieving high transferability. Additionally, we provide some explanatory insights regarding our method and the effect of optimizing for adversarial examples using intermediate feature maps.
Original language | English |
---|---|
Journal | Proceedings of the IEEE International Conference on Computer Vision |
Pages (from-to) | 4732-4741 |
Number of pages | 10 |
ISSN | 1550-5499 |
DOIs | |
Publication status | Published - Oct 2019 |
Event | 17th IEEE/CVF International Conference on Computer Vision, ICCV 2019 - Seoul, Korea, Republic of Duration: 27 Oct 2019 → 2 Nov 2019 |
Conference
Conference | 17th IEEE/CVF International Conference on Computer Vision, ICCV 2019 |
---|---|
Country | Korea, Republic of |
City | Seoul |
Period | 27/10/2019 → 02/11/2019 |
Sponsor | Computer Vision Foundation, IEEE |
Bibliographical note
Publisher Copyright:
© 2019 IEEE.
ID: 301824085