Diverse Generative Adversarial Perturbations on Attention Space for Transferable Adversarial Attacks

08/11/2022
by   Woo Jae Kim, et al.
0

Adversarial attacks with improved transferability - the ability of an adversarial example crafted on a known model to also fool unknown models - have recently received much attention due to their practicality. Nevertheless, existing transferable attacks craft perturbations in a deterministic manner and often fail to fully explore the loss surface, thus falling into a poor local optimum and suffering from low transferability. To solve this problem, we propose Attentive-Diversity Attack (ADA), which disrupts diverse salient features in a stochastic manner to improve transferability. Primarily, we perturb the image attention to disrupt universal features shared by different models. Then, to effectively avoid poor local optima, we disrupt these features in a stochastic manner and explore the search space of transferable perturbations more exhaustively. More specifically, we use a generator to produce adversarial perturbations that each disturbs features in different ways depending on an input latent code. Extensive experimental evaluations demonstrate the effectiveness of our method, outperforming the transferability of state-of-the-art methods. Codes are available at https://github.com/wkim97/ADA.

READ FULL TEXT
research
09/08/2022

Incorporating Locality of Images to Generate Targeted Transferable Adversarial Examples

Despite that leveraging the transferability of adversarial examples can ...
research
08/08/2023

PAIF: Perception-Aware Infrared-Visible Image Fusion for Attack-Tolerant Semantic Segmentation

Infrared and visible image fusion is a powerful technique that combines ...
research
10/03/2020

A Deep Genetic Programming based Methodology for Art Media Classification Robust to Adversarial Perturbations

Art Media Classification problem is a current research area that has att...
research
07/24/2022

Proving Common Mechanisms Shared by Twelve Methods of Boosting Adversarial Transferability

Although many methods have been proposed to enhance the transferability ...
research
04/08/2020

Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking

The success of DNNs has driven the extensive applications of person re-i...
research
10/08/2020

A Unified Approach to Interpreting and Boosting Adversarial Transferability

In this paper, we use the interaction inside adversarial perturbations t...
research
08/27/2021

Disrupting Adversarial Transferability in Deep Neural Networks

Adversarial attack transferability is a well-recognized phenomenon in de...

Please sign up or login with your details

Forgot password? Click here to reset