Embodied Active Defense: Leveraging Recurrent Feedback to Counter Adversarial Patches

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wu, Lingxuan, Yang, Xiao, Dong, Yinpeng, Xie, Liuwei, Su, Hang, Zhu, Jun
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909155672981504
author Wu, Lingxuan
Yang, Xiao
Dong, Yinpeng
Xie, Liuwei
Su, Hang
Zhu, Jun
author_facet Wu, Lingxuan
Yang, Xiao
Dong, Yinpeng
Xie, Liuwei
Su, Hang
Zhu, Jun
contents The vulnerability of deep neural networks to adversarial patches has motivated numerous defense strategies for boosting model robustness. However, the prevailing defenses depend on single observation or pre-established adversary information to counter adversarial patches, often failing to be confronted with unseen or adaptive adversarial attacks and easily exhibiting unsatisfying performance in dynamic 3D environments. Inspired by active human perception and recurrent feedback mechanisms, we develop Embodied Active Defense (EAD), a proactive defensive strategy that actively contextualizes environmental information to address misaligned adversarial patches in 3D real-world settings. To achieve this, EAD develops two central recurrent sub-modules, i.e., a perception module and a policy module, to implement two critical functions of active vision. These models recurrently process a series of beliefs and observations, facilitating progressive refinement of their comprehension of the target object and enabling the development of strategic actions to counter adversarial patches in 3D environments. To optimize learning efficiency, we incorporate a differentiable approximation of environmental dynamics and deploy patches that are agnostic to the adversary strategies. Extensive experiments demonstrate that EAD substantially enhances robustness against a variety of patches within just a few steps through its action policy in safety-critical tasks (e.g., face recognition and object detection), without compromising standard accuracy. Furthermore, due to the attack-agnostic characteristic, EAD facilitates excellent generalization to unseen attacks, diminishing the averaged attack success rate by 95 percent across a range of unseen adversarial attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2404_00540
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Embodied Active Defense: Leveraging Recurrent Feedback to Counter Adversarial Patches
Wu, Lingxuan
Yang, Xiao
Dong, Yinpeng
Xie, Liuwei
Su, Hang
Zhu, Jun
Computer Vision and Pattern Recognition
Artificial Intelligence
The vulnerability of deep neural networks to adversarial patches has motivated numerous defense strategies for boosting model robustness. However, the prevailing defenses depend on single observation or pre-established adversary information to counter adversarial patches, often failing to be confronted with unseen or adaptive adversarial attacks and easily exhibiting unsatisfying performance in dynamic 3D environments. Inspired by active human perception and recurrent feedback mechanisms, we develop Embodied Active Defense (EAD), a proactive defensive strategy that actively contextualizes environmental information to address misaligned adversarial patches in 3D real-world settings. To achieve this, EAD develops two central recurrent sub-modules, i.e., a perception module and a policy module, to implement two critical functions of active vision. These models recurrently process a series of beliefs and observations, facilitating progressive refinement of their comprehension of the target object and enabling the development of strategic actions to counter adversarial patches in 3D environments. To optimize learning efficiency, we incorporate a differentiable approximation of environmental dynamics and deploy patches that are agnostic to the adversary strategies. Extensive experiments demonstrate that EAD substantially enhances robustness against a variety of patches within just a few steps through its action policy in safety-critical tasks (e.g., face recognition and object detection), without compromising standard accuracy. Furthermore, due to the attack-agnostic characteristic, EAD facilitates excellent generalization to unseen attacks, diminishing the averaged attack success rate by 95 percent across a range of unseen adversarial attacks.
title Embodied Active Defense: Leveraging Recurrent Feedback to Counter Adversarial Patches
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2404.00540