Cooperative Backdoor Attack in Decentralized Reinforcement Learning with Theoretical Guarantee.
Mengtong GaoYifei ZouZuyuan ZhangXiuzhen ChengDongxiao YuPublished in: CoRR (2024)
Keyphrases
- cooperative
- theoretical guarantees
- reinforcement learning
- multi agent
- multiagent reinforcement learning
- policy iteration
- worst case
- function approximation
- state space
- model free
- multi agent reinforcement learning
- markov decision processes
- reinforcement learning algorithms
- machine learning
- dynamic programming
- multi agent systems
- secret key
- optimal policy
- action selection
- learning algorithm
- countermeasures
- game theory
- temporal difference
- similarity search
- peer to peer
- least squares
- fully distributed
- evolutionary learning
- cost function
- optimal solution
- abstract argumentation