Security and privacy issues in deep reinforcement learning: Threats and countermeasures

K Mo, P Ye, X Ren, S Wang, W Li, J Li - ACM Computing Surveys, 2024 - dl.acm.org
Deep Reinforcement Learning (DRL) is an essential subfield of Artificial Intelligence (AI),
where agents interact with environments to learn policies for solving complex tasks. In recent …

Adversarial Machine Learning Attacks and Defences in Multi-Agent Reinforcement Learning

M Standen, J Kim, C Szabo - ACM Computing Surveys, 2025 - dl.acm.org
Multi-Agent Reinforcement Learning (MARL) is susceptible to Adversarial Machine Learning
(AML) attacks. Execution-time AML attacks against MARL are complex due to effects that …

Robust cooperative multi-agent reinforcement learning via multi-view message certification

L Yuan, T Jiang, L Li, F Chen, Z Zhang, Y Yu - Science China Information …, 2024 - Springer
Many multi-agent scenarios require message sharing among agents to promote
coordination, hastening the robustness of multi-agent communication when policies are …

Efficient multi-agent communication via self-supervised information aggregation

C Guan, F Chen, L Yuan, C Wang… - Advances in …, 2022 - proceedings.neurips.cc
Utilizing messages from teammates can improve coordination in cooperative Multi-agent
Reinforcement Learning (MARL). To obtain meaningful information for decision-making …

What is the solution for state-adversarial multi-agent reinforcement learning?

S Han, S Su, S He, S Han, H Yang, S Zou… - arxiv preprint arxiv …, 2022 - arxiv.org
Various methods for Multi-Agent Reinforcement Learning (MARL) have been developed
with the assumption that agents' policies are based on accurate state information. However …

Instructed diffuser with temporal condition guidance for offline reinforcement learning

J Hu, Y Sun, S Huang, SY Guo, H Chen, L Shen… - arxiv preprint arxiv …, 2023 - arxiv.org
Recent works have shown the potential of diffusion models in computer vision and natural
language processing. Apart from the classical supervised learning fields, diffusion models …

Certified policy smoothing for cooperative multi-agent reinforcement learning

R Mu, W Ruan, LS Marcolino, G **, Q Ni - Proceedings of the AAAI …, 2023 - ojs.aaai.org
Cooperative multi-agent reinforcement learning (c-MARL) is widely applied in safety-critical
scenarios, thus the analysis of robustness for c-MARL models is profoundly important …

Rethinking adversarial policies: a generalized attack formulation and provable defense in RL

X Liu, S Chakraborty, Y Sun, F Huang - arxiv preprint arxiv:2305.17342, 2023 - arxiv.org
Most existing works focus on direct perturbations to the victim's state/action or the underlying
transition dynamics to demonstrate the vulnerability of reinforcement learning agents to …

Robust multi-agent communication via multi-view message certification

L Yuan, T Jiang, L Li, F Chen, Z Zhang, Y Yu - arxiv preprint arxiv …, 2023 - arxiv.org
Many multi-agent scenarios require message sharing among agents to promote
coordination, hastening the robustness of multi-agent communication when policies are …

Optimal cost constrained adversarial attacks for multiple agent systems

Z Lu, G Liu, L Lai, W Xu - 2024 58th Annual Conference on …, 2024 - ieeexplore.ieee.org
Since many security-related applications use multi-agent reinforcement learning as their
underlying algorithms, the study on the adversarial attacks against mutli-agent reinforcement …