EUVIMEDEUVIMED entdecken
Crossref

Adversarial Attack on Black-Box Multi-Agent by Adaptive Perturbation

Jianming Chen, Yawen Wang, Junjie Wang, Xiaofei Xie, Yuanzhe Hu, Qing Wang, Fanjiang Xu

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (35) · 29359-29367

DOI: 10.1609/aaai.v40i35.40176 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Jianming Chen und 6 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Evaluating security and reliability for multi-agent systems (MAS) is urgent as they become increasingly prevalent in various applications. As an evaluation technique, existing adversarial attack frameworks face certain limitations, e.g., impracticality due to the requirement of white-box information or high control authority, and a lack of stealthiness or effectiveness as they often target all agents or specific fixed agents. To address these issues, we propose AdapAM, a novel framework for adversarial attacks on black-box MAS. AdapAM incorporates two key components: (1) Adaptive Selection Policy simultaneously selects the victim and determines the anticipated malicious action (the action would lead to the worst impact on MAS), balancing effectiveness and stealthiness. (2) Proxy-based Perturbation to Induce Malicious Action utilizes generative adversarial imitation learning to approximate the target MAS, allowing AdapAM to generate perturbed observations using white-box information and thus induce victims to execute malicious action in black-box settings. We evaluate AdapAM across eight multi-agent environments and compare it with four state-of-the-art and commonly-used baselines. Results demonstrate that AdapAM achieves the best attack performance in different perturbation rates. Besides, AdapAM-generated perturbations are the least noisy and hardest to detect, emphasizing the stealthiness.

Zitieren

APA

Chen, J., Wang, Y., Wang, J., Xie, X., Hu, Y., Wang, Q., & Xu, F. (2026). Adversarial Attack on Black-Box Multi-Agent by Adaptive Perturbation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(35), 29359-29367. https://doi.org/10.1609/aaai.v40i35.40176

Vancouver

Chen J, Wang Y, Wang J, Xie X, Hu Y, Wang Q, et al. Adversarial Attack on Black-Box Multi-Agent by Adaptive Perturbation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(35):29359-29367. doi:10.1609/aaai.v40i35.40176

BibTeX
@article{Chen2026,
  author = {Chen, Jianming and Wang, Yawen and Wang, Junjie and Xie, Xiaofei and Hu, Yuanzhe and Wang, Qing and Xu, Fanjiang},
  title = {Adversarial Attack on Black-Box Multi-Agent by Adaptive Perturbation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {35},
  pages = {29359--29367},
  doi = {10.1609/aaai.v40i35.40176}
}
Kostenlos starten →