EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TReS‐BD: Trigger‐Aware Reward Shaping for Efficient and Stealthy Backdoor Attacks in Cooperative Multi‐Agent RL

Chaoyang Zhu, Xiao Zhu, Xiangcheng Li, Tuanfa Qin

IET Information Security · 2026-01 · 2026 (1)

DOI: 10.1049/ise2/3609361 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IET Information Security, online erschienen am 23.02.2026. Verfasst von Chaoyang Zhu und 3 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cooperative multi‐agent reinforcement learning (CMARL) has become a powerful paradigm for applications in autonomous driving, smart grids, and distributed robotics. However, its increasing adoption in safety‐critical scenarios raises severe concerns about security vulnerabilities, particularly backdoor attacks, where malicious behaviors are embedded and activated only under specific triggers. This paper presents TReS‐BD, a trigger‐aware reward shaping framework that systematically exploits CMARL vulnerabilities at the reward level. By formulating backdoor manipulation as a Bayes‐Adaptive Markov Decision Process (BAMDP), TReS‐BD treats trigger conditions as latent variables and integrates KL‐regularized policy optimization to maintain nominal task performance while inducing adversarial behaviors under triggers. Extensive experiments on standard benchmarks, including multi‐agent particle environments (MPE) and StarCraft Multi‐Agent Challenge (SMAC), with representative algorithms MADDPG and MAPPO, demonstrate that TReS‐BD achieves over 90% attack success with less than 1% training data poisoning, while maintaining normal performance and evading current detection mechanisms. Furthermore, a systematic analysis of existing defenses reveals their ineffectiveness against such reward‐level attacks, underscoring a critical blind spot in current CMARL security frameworks. Our findings highlight the urgent need for robust defense strategies and provide new insights into safeguarding cooperative multi‐agent systems against stealthy and efficient backdoor threats.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Zhu, C., Zhu, X., Li, X., & Qin, T. (2026). TReS‐BD: Trigger‐Aware Reward Shaping for Efficient and Stealthy Backdoor Attacks in Cooperative Multi‐Agent RL. IET Information Security, 2026(1). https://doi.org/10.1049/ise2/3609361

Vancouver

Zhu C, Zhu X, Li X, Qin T. TReS‐BD: Trigger‐Aware Reward Shaping for Efficient and Stealthy Backdoor Attacks in Cooperative Multi‐Agent RL. IET Information Security. 2026;2026(1). doi:10.1049/ise2/3609361

BibTeX
@article{Zhu2026,
  author = {Zhu, Chaoyang and Zhu, Xiao and Li, Xiangcheng and Qin, Tuanfa},
  title = {TReS‐BD: Trigger‐Aware Reward Shaping for Efficient and Stealthy Backdoor Attacks in Cooperative Multi‐Agent RL},
  journal = {IET Information Security},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1049/ise2/3609361}
}
Kostenlos starten →