MARNET: Backdoor Attacks against Value-Decomposition Multi-Agent Reinforcement Learning

no code implementations • 29 Sep 2021 • Yanjiao Chen, Zhicong Zheng, Xueluan Gong

Recent works have revealed that backdoor attacks against Deep Reinforcement Learning (DRL) could lead to abnormal action selection of the agent, which may result in failure or even catastrophe in crucial decision processes.

Backdoor Attack reinforcement-learning +4